One read of Claude Developer Platformapi-20261001T213726Z
336 pages moved out of 748 read.
Pages moved
336
significant first
Pages read
748
in this capture
Captured
21:37 UTC
Corpus hash
6a28c636443e
index-hash
What this read moved
126-150 of 336, page 6 of 14This capture is too large to show at once. Changes 126-150 of 336 are below, significant first; the rest are on the following screens.
api/beta/tunnels/certificates Changed · +8 / -0 lines
from line 125
125125
126126 - `"ce-plugins-2026-09-01"`
127127
128 - `"spend-limit-reads-2026-09-26"`
129
128130- `"anthropic-workspace-id": optional string`
129131
130132 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 330
328330
329331 - `"ce-plugins-2026-09-01"`
330332
333 - `"spend-limit-reads-2026-09-26"`
334
331335- `"anthropic-workspace-id": optional string`
332336
333337 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 535
531535
532536 - `"ce-plugins-2026-09-01"`
533537
538 - `"spend-limit-reads-2026-09-26"`
539
534540- `"anthropic-workspace-id": optional string`
535541
536542 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 736
730736 - `"mcp-client-2026-09-15"`
731737
732738 - `"ce-plugins-2026-09-01"`
739
740 - `"spend-limit-reads-2026-09-26"`
733741
734742- `"anthropic-workspace-id": optional string`
735743
api/beta/user_profiles Changed · +10 / -0 lines
from line 117
117117
118118 - `"ce-plugins-2026-09-01"`
119119
120 - `"spend-limit-reads-2026-09-26"`
121
120122- `"anthropic-workspace-id": optional string`
121123
122124 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 542
540542
541543 - `"ce-plugins-2026-09-01"`
542544
545 - `"spend-limit-reads-2026-09-26"`
546
543547- `"anthropic-workspace-id": optional string`
544548
545549 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 840
836840
837841 - `"ce-plugins-2026-09-01"`
838842
843 - `"spend-limit-reads-2026-09-26"`
844
839845- `"anthropic-workspace-id": optional string`
840846
841847 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1131
11251131
11261132 - `"ce-plugins-2026-09-01"`
11271133
1134 - `"spend-limit-reads-2026-09-26"`
1135
11281136- `"anthropic-workspace-id": optional string`
11291137
11301138 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1521
15131521 - `"mcp-client-2026-09-15"`
15141522
15151523 - `"ce-plugins-2026-09-01"`
1524
1525 - `"spend-limit-reads-2026-09-26"`
15161526
15171527- `"anthropic-workspace-id": optional string`
15181528
api/beta/vaults Changed · +26 / -0 lines
from line 117
117117
118118 - `"ce-plugins-2026-09-01"`
119119
120 - `"spend-limit-reads-2026-09-26"`
121
120122- `"anthropic-workspace-id": optional string`
121123
122124 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 335
333335
334336 - `"ce-plugins-2026-09-01"`
335337
338 - `"spend-limit-reads-2026-09-26"`
339
336340- `"anthropic-workspace-id": optional string`
337341
338342 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 533
529533
530534 - `"ce-plugins-2026-09-01"`
531535
536 - `"spend-limit-reads-2026-09-26"`
537
532538- `"anthropic-workspace-id": optional string`
533539
534540 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 722
716722
717723 - `"ce-plugins-2026-09-01"`
718724
725 - `"spend-limit-reads-2026-09-26"`
726
719727- `"anthropic-workspace-id": optional string`
720728
721729 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 930
922930
923931 - `"ce-plugins-2026-09-01"`
924932
933 - `"spend-limit-reads-2026-09-26"`
934
925935- `"anthropic-workspace-id": optional string`
926936
927937 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1087
10771087
10781088 - `"ce-plugins-2026-09-01"`
10791089
1090 - `"spend-limit-reads-2026-09-26"`
1091
10801092- `"anthropic-workspace-id": optional string`
10811093
10821094 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1331
13191331
13201332 - `"ce-plugins-2026-09-01"`
13211333
1334 - `"spend-limit-reads-2026-09-26"`
1335
13221336- `"anthropic-workspace-id": optional string`
13231337
13241338 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1843
18291843
18301844 - `"ce-plugins-2026-09-01"`
18311845
1846 - `"spend-limit-reads-2026-09-26"`
1847
18321848- `"anthropic-workspace-id": optional string`
18331849
18341850 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 2166
21502166
21512167 - `"ce-plugins-2026-09-01"`
21522168
2169 - `"spend-limit-reads-2026-09-26"`
2170
21532171- `"anthropic-workspace-id": optional string`
21542172
21552173 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 2480
24622480
24632481 - `"ce-plugins-2026-09-01"`
24642482
2483 - `"spend-limit-reads-2026-09-26"`
2484
24652485- `"anthropic-workspace-id": optional string`
24662486
24672487 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 2933
29132933
29142934 - `"ce-plugins-2026-09-01"`
29152935
2936 - `"spend-limit-reads-2026-09-26"`
2937
29162938- `"anthropic-workspace-id": optional string`
29172939
29182940 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 3094
30723094
30733095 - `"ce-plugins-2026-09-01"`
30743096
3097 - `"spend-limit-reads-2026-09-26"`
3098
30753099- `"anthropic-workspace-id": optional string`
30763100
30773101 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 3408
33843408 - `"mcp-client-2026-09-15"`
33853409
33863410 - `"ce-plugins-2026-09-01"`
3411
3412 - `"spend-limit-reads-2026-09-26"`
33873413
33883414- `"anthropic-workspace-id": optional string`
33893415
api/beta/vaults/credentials Changed · +14 / -0 lines
from line 123
123123
124124 - `"ce-plugins-2026-09-01"`
125125
126 - `"spend-limit-reads-2026-09-26"`
127
126128- `"anthropic-workspace-id": optional string`
127129
128130 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 635
633635
634636 - `"ce-plugins-2026-09-01"`
635637
638 - `"spend-limit-reads-2026-09-26"`
639
636640- `"anthropic-workspace-id": optional string`
637641
638642 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 958
954958
955959 - `"ce-plugins-2026-09-01"`
956960
961 - `"spend-limit-reads-2026-09-26"`
962
957963- `"anthropic-workspace-id": optional string`
958964
959965 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1272
12661272
12671273 - `"ce-plugins-2026-09-01"`
12681274
1275 - `"spend-limit-reads-2026-09-26"`
1276
12691277- `"anthropic-workspace-id": optional string`
12701278
12711279 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1725
17171725
17181726 - `"ce-plugins-2026-09-01"`
17191727
1728 - `"spend-limit-reads-2026-09-26"`
1729
17201730- `"anthropic-workspace-id": optional string`
17211731
17221732 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 1886
18761886
18771887 - `"ce-plugins-2026-09-01"`
18781888
1889 - `"spend-limit-reads-2026-09-26"`
1890
18791891- `"anthropic-workspace-id": optional string`
18801892
18811893 Optional header to select the Workspace for this request. The value is a Workspace ID (for example, `wrkspc_011CZkZaBF1tNoB5wlCeusgy`).
from line 2200
21882200 - `"mcp-client-2026-09-15"`
21892201
21902202 - `"ce-plugins-2026-09-01"`
2203
2204 - `"spend-limit-reads-2026-09-26"`
21912205
21922206- `"anthropic-workspace-id": optional string`
21932207
api/completions Changed · +32 / -18 lines
from line 129
129129
130130 - `"ce-plugins-2026-09-01"`
131131
132 - `"spend-limit-reads-2026-09-26"`
133
132134### Body parameters
133135
134136- `max_tokens_to_sample: number`
from line 211
209211
210212 Powerful intelligence for long-running agents and coding
211213
214 - `"claude-mythos-preview"`
215
216 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
217
218 New class of intelligence, strongest in coding and cybersecurity
219
212220 - `"claude-sonnet-4-5"`
213221
222 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
223
214224 High-performance model for agents and coding
215225
216226 - `"claude-sonnet-4-5-20250929"`
217227
228 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
229
218230 High-performance model for agents and coding
219231
220 - `"claude-mythos-preview"`
221
222 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
223
224 New class of intelligence, strongest in coding and cybersecurity
225
226232 - `string`
227233
228234- `prompt: string`
from line 407
401407
402408 Powerful intelligence for long-running agents and coding
403409
410 - `"claude-mythos-preview"`
411
412 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
413
414 New class of intelligence, strongest in coding and cybersecurity
415
404416 - `"claude-sonnet-4-5"`
405417
418 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
419
406420 High-performance model for agents and coding
407421
408422 - `"claude-sonnet-4-5-20250929"`
409423
424 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
425
410426 High-performance model for agents and coding
411427
412 - `"claude-mythos-preview"`
413
414 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
415
416 New class of intelligence, strongest in coding and cybersecurity
417
418428 - `string`
419429
420430 - `stop_reason: string or null`
from line 560
550560
551561 Powerful intelligence for long-running agents and coding
552562
563 - `"claude-mythos-preview"`
564
565 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
566
567 New class of intelligence, strongest in coding and cybersecurity
568
553569 - `"claude-sonnet-4-5"`
554570
571 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
572
555573 High-performance model for agents and coding
556574
557575 - `"claude-sonnet-4-5-20250929"`
558576
559 High-performance model for agents and coding
577 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
560578
561 - `"claude-mythos-preview"`
562
563 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
564
565 New class of intelligence, strongest in coding and cybersecurity
579 High-performance model for agents and coding
566580
567581 - `string`
568582
api/completions/create Changed · +22 / -12 lines
from line 127
127127
128128 - `"ce-plugins-2026-09-01"`
129129
130 - `"spend-limit-reads-2026-09-26"`
131
130132## Body parameters
131133
132134- `max_tokens_to_sample: number`
from line 209
207209
208210 Powerful intelligence for long-running agents and coding
209211
212 - `"claude-mythos-preview"`
213
214 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
215
216 New class of intelligence, strongest in coding and cybersecurity
217
210218 - `"claude-sonnet-4-5"`
211219
220 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
221
212222 High-performance model for agents and coding
213223
214224 - `"claude-sonnet-4-5-20250929"`
215225
226 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
227
216228 High-performance model for agents and coding
217229
218 - `"claude-mythos-preview"`
219
220 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
221
222 New class of intelligence, strongest in coding and cybersecurity
223
224230 - `string`
225231
226232- `prompt: string`
from line 405
399405
400406 Powerful intelligence for long-running agents and coding
401407
408 - `"claude-mythos-preview"`
409
410 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
411
412 New class of intelligence, strongest in coding and cybersecurity
413
402414 - `"claude-sonnet-4-5"`
403415
416 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
417
404418 High-performance model for agents and coding
405419
406420 - `"claude-sonnet-4-5-20250929"`
407421
408 High-performance model for agents and coding
422 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
409423
410 - `"claude-mythos-preview"`
411
412 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
413
414 New class of intelligence, strongest in coding and cybersecurity
424 High-performance model for agents and coding
415425
416426 - `string`
417427
api/compliance Changed · +21 / -1 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
Nothing in the body moved in this read. What changed is above.
api/compliance/activities Changed · +42 / -2 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
Nothing in the body moved in this read. What changed is above.
api/compliance/activities/list Changed · +21 / -1 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
Nothing in the body moved in this read. What changed is above.
api/messages Changed · +90 / -54 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
from line 1139
11391139
11401140 Powerful intelligence for long-running agents and coding
11411141
1142 - `"claude-mythos-preview"`
1143
1144 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1145
1146 New class of intelligence, strongest in coding and cybersecurity
1147
11421148 - `"claude-sonnet-4-5"`
11431149
1150 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1151
11441152 High-performance model for agents and coding
11451153
11461154 - `"claude-sonnet-4-5-20250929"`
11471155
1156 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1157
11481158 High-performance model for agents and coding
11491159
1150 - `"claude-mythos-preview"`
1151
1152 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1153
1154 New class of intelligence, strongest in coding and cybersecurity
1155
11561160 - `string`
11571161
11581162- `cache_control: optional CacheControlEphemeral or null`
from line 3984
39803984
39813985 Powerful intelligence for long-running agents and coding
39823986
3987 - `"claude-mythos-preview"`
3988
3989 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
3990
3991 New class of intelligence, strongest in coding and cybersecurity
3992
39833993 - `"claude-sonnet-4-5"`
39843994
3995 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
3996
39853997 High-performance model for agents and coding
39863998
39873999 - `"claude-sonnet-4-5-20250929"`
39884000
4001 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
4002
39894003 High-performance model for agents and coding
39904004
3991 - `"claude-mythos-preview"`
3992
3993 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
3994
3995 New class of intelligence, strongest in coding and cybersecurity
3996
39974005 - `string`
39984006
39994007 - `role: "assistant"`
from line 5630
56225630
56235631 Powerful intelligence for long-running agents and coding
56245632
5633 - `"claude-mythos-preview"`
5634
5635 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
5636
5637 New class of intelligence, strongest in coding and cybersecurity
5638
56255639 - `"claude-sonnet-4-5"`
56265640
5641 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
5642
56275643 High-performance model for agents and coding
56285644
56295645 - `"claude-sonnet-4-5-20250929"`
56305646
5647 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
5648
56315649 High-performance model for agents and coding
56325650
5633 - `"claude-mythos-preview"`
5634
5635 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
5636
5637 New class of intelligence, strongest in coding and cybersecurity
5638
56395651 - `string`
56405652
56415653- `cache_control: optional CacheControlEphemeral or null`
from line 8407
83958407
83968408 - `url: string`
83978409
8398 The final post-redirect URL the download was served from.
8399
8400 maxLength: 4096, pattern: ^[^\x00-\x1f\x7f-\x9f\u2028\u2029]*$
8401
8402 - `path: optional string or null`
8403
8404 Where the executor saved the file, on the executor's filesystem. Only included when another tool in the same environment can read the file at that path.
8405
8406 maxLength: 4096, pattern: ^[^\x00-\x1f\x7f-\x9f\u2028\u2029]*$
8407
8408 - `size_bytes: optional number or null`
8409
8410 The completed download's size.
8411
8412 minimum: 0
8413
8414### Browser State Change Download Failed
8415
8416- `BrowserStateChangeDownloadFailed object`
8417
8418 A file download that failed — or was cancelled — during this call.
8419
8420 - `type: "download_failed"`
8421
8422 - `download_id: string`
8423
8424 The caller-assigned identifier for this download, stable across the state changes reporting it.
8425
8426 minLength: 1, maxLength: 4096, pattern: ^[^\x00-\x1f\x7f-\x9f\u2028\u2029]*$
8427
8428 - `url: string`
8429
8430 The final post-redirect URL the download was served from.
8431
8432 maxLength: 4096, pattern: ^[^\x00-\x1f\x7f-\x9f\u2028\u2029]*$
8433
8434 - `error: optional string or null`
8435
8436 The failure or cancellation detail, when known.
8437
8438 maxLength: 4096, pattern: ^[^\x00-\x1f\x7f-\x9f\u2028\u2029]*$
8439
8410 The final post-redire
api/messages/batches Changed · +50 / -30 lines
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
from line 1159
11591159
11601160 Powerful intelligence for long-running agents and coding
11611161
1162 - `"claude-mythos-preview"`
1163
1164 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1165
1166 New class of intelligence, strongest in coding and cybersecurity
1167
11621168 - `"claude-sonnet-4-5"`
11631169
1170 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1171
11641172 High-performance model for agents and coding
11651173
11661174 - `"claude-sonnet-4-5-20250929"`
11671175
1176 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1177
11681178 High-performance model for agents and coding
11691179
1170 - `"claude-mythos-preview"`
1171
1172 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1173
1174 New class of intelligence, strongest in coding and cybersecurity
1175
11761180 - `string`
11771181
11781182 - `cache_control: optional CacheControlEphemeral or null`
from line 4790
47864790
47874791 Powerful intelligence for long-running agents and coding
47884792
4793 - `"claude-mythos-preview"`
4794
4795 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
4796
4797 New class of intelligence, strongest in coding and cybersecurity
4798
47894799 - `"claude-sonnet-4-5"`
47904800
4801 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
4802
47914803 High-performance model for agents and coding
47924804
47934805 - `"claude-sonnet-4-5-20250929"`
47944806
4807 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
4808
47954809 High-performance model for agents and coding
47964810
4797 - `"claude-mythos-preview"`
4798
4799 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
4800
4801 New class of intelligence, strongest in coding and cybersecurity
4802
48034811 - `string`
48044812
48054813 - `role: "assistant"`
from line 6303
62956303
62966304 Powerful intelligence for long-running agents and coding
62976305
6306 - `"claude-mythos-preview"`
6307
6308 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
6309
6310 New class of intelligence, strongest in coding and cybersecurity
6311
62986312 - `"claude-sonnet-4-5"`
62996313
6314 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
6315
63006316 High-performance model for agents and coding
63016317
63026318 - `"claude-sonnet-4-5-20250929"`
63036319
6320 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
6321
63046322 High-performance model for agents and coding
63056323
6306 - `"claude-mythos-preview"`
6307
6308 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
6309
6310 New class of intelligence, strongest in coding and cybersecurity
6311
63126324 - `string`
63136325
63146326 - `role: "assistant"`
from line 7588
75767588
75777589 Powerful intelligence for long-running agents and coding
75787590
7591 - `"claude-mythos-preview"`
7592
7593 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
7594
7595 New class of intelligence, strongest in coding and cybersecurity
7596
75797597 - `"claude-sonnet-4-5"`
75807598
7599 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
7600
75817601 High-performance model for agents and coding
75827602
75837603 - `"claude-sonnet-4-5-20250929"`
75847604
7605 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
7606
75857607 High-performance model for agents and coding
75867608
7587 - `"claude-mythos-preview"`
7588
7589 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
7590
7591 New class of intelligence, strongest in coding and cybersecurity
7592
75937609 - `string`
75947610
75957611 - `role: "assistant"`
from line 8701
86858701
86868702 - `cache_miss_reason: CacheMissReason or null`
86878703
8688 Explains why the prompt cache could not fully reuse the prefix from the request identified by `diagnostics.previous_message_id`. `null` means diagnosis is still pending — the response was serialized before the background comparison completed.
8689
8690 - `CacheMissModelChanged object`
8691
8692 - `type: "model_changed"`
8693
8694 default: model_changed
8695
8696 - `cache_missed_input_tokens: number`
8697
8698 Approximate number of input tokens that would have been read from cache had the prefix matched the previous request.
8699
8700 - `CacheMissSystemChanged object`
8701
8702 - `type: "system_changed"`
8703
8704 default: system_changed
8705
8706 - `cache_missed_input_tokens: number`
8707
8708 Approximate number of input tokens that would have been read from cache had the prefix matched the previous request.
8709
8710 - `CacheMissToolsChanged object`
8711
8712 - `type: "tools_changed"`
8713
8714 default: tools_changed
8715
8716 - `cache_missed_input_tokens: number`
8717
8718 Approximate number of input tokens that would have been read from cache had the prefix matched the previous request.
8719
8720 - `CacheMissMessagesChanged object`
8721
8722 - `type: "messages_changed"`
8723
8724 default: messages_changed
8725
8726 - `cache_missed_input_tokens: number`
8727
8728 Approximate number of input tokens that would have been read from cache had the prefix matched the previous request.
8729
8730 - `CacheMissPreviousMessageNotFound object`
8731
8732 - `type: "previous_message_not_found"`
8733
8734 default: previous_message_not_found
8735
8736 - `CacheMissUnavailable object`
8737
8738 - `type: "unavailable"`
8739
8740 default: unavailable
8741
8742 - `model: Model`
8743
8744 The model that will complete your prompt.
8745
8746 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
8704 Explains why the prompt cache could not fully reuse the prefix from the request identified by `diagnostics.previous_message_id`. `null` means diagnosis is still pending — the response was serialized before the background comparison complet
api/messages/batches/create Changed · +10 / -6 lines
from line 1157
11571157
11581158 Powerful intelligence for long-running agents and coding
11591159
1160 - `"claude-mythos-preview"`
1161
1162 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1163
1164 New class of intelligence, strongest in coding and cybersecurity
1165
11601166 - `"claude-sonnet-4-5"`
11611167
1168 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1169
11621170 High-performance model for agents and coding
11631171
11641172 - `"claude-sonnet-4-5-20250929"`
11651173
1166 High-performance model for agents and coding
1174 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
11671175
1168 - `"claude-mythos-preview"`
1169
1170 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1171
1172 New class of intelligence, strongest in coding and cybersecurity
1176 High-performance model for agents and coding
11731177
11741178 - `string`
11751179
api/messages/batches/results Changed · +10 / -6 lines
from line 952
952952
953953 Powerful intelligence for long-running agents and coding
954954
955 - `"claude-mythos-preview"`
956
957 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
958
959 New class of intelligence, strongest in coding and cybersecurity
960
955961 - `"claude-sonnet-4-5"`
956962
963 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
964
957965 High-performance model for agents and coding
958966
959967 - `"claude-sonnet-4-5-20250929"`
960968
961 High-performance model for agents and coding
969 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
962970
963 - `"claude-mythos-preview"`
964
965 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
966
967 New class of intelligence, strongest in coding and cybersecurity
971 High-performance model for agents and coding
968972
969973 - `string`
970974
api/messages/count_tokens Changed · +10 / -6 lines
from line 1125
11251125
11261126 Powerful intelligence for long-running agents and coding
11271127
1128 - `"claude-mythos-preview"`
1129
1130 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1131
1132 New class of intelligence, strongest in coding and cybersecurity
1133
11281134 - `"claude-sonnet-4-5"`
11291135
1136 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1137
11301138 High-performance model for agents and coding
11311139
11321140 - `"claude-sonnet-4-5-20250929"`
11331141
1134 High-performance model for agents and coding
1142 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
11351143
1136 - `"claude-mythos-preview"`
1137
1138 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1139
1140 New class of intelligence, strongest in coding and cybersecurity
1144 High-performance model for agents and coding
11411145
11421146 - `string`
11431147
api/messages/create Changed · +20 / -12 lines
from line 1137
11371137
11381138 Powerful intelligence for long-running agents and coding
11391139
1140 - `"claude-mythos-preview"`
1141
1142 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1143
1144 New class of intelligence, strongest in coding and cybersecurity
1145
11401146 - `"claude-sonnet-4-5"`
11411147
1148 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1149
11421150 High-performance model for agents and coding
11431151
11441152 - `"claude-sonnet-4-5-20250929"`
11451153
1154 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1155
11461156 High-performance model for agents and coding
11471157
1148 - `"claude-mythos-preview"`
1149
1150 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
1151
1152 New class of intelligence, strongest in coding and cybersecurity
1153
11541158 - `string`
11551159
11561160- `cache_control: optional CacheControlEphemeral or null`
from line 3982
39783982
39793983 Powerful intelligence for long-running agents and coding
39803984
3985 - `"claude-mythos-preview"`
3986
3987 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
3988
3989 New class of intelligence, strongest in coding and cybersecurity
3990
39813991 - `"claude-sonnet-4-5"`
39823992
3993 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
3994
39833995 High-performance model for agents and coding
39843996
39853997 - `"claude-sonnet-4-5-20250929"`
39863998
3987 High-performance model for agents and coding
3999 **Deprecated**: Will reach end-of-life on November 30, 2026. Please migrate to claude-sonnet-5-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
39884000
3989 - `"claude-mythos-preview"`
3990
3991 **Deprecated**: Will reach end-of-life on June 30, 2026. Please migrate to claude-mythos-5. Visit https://docs.anthropic.com/en/docs/resources/model-deprecations for more information.
3992
3993 New class of intelligence, strongest in coding and cybersecurity
4001 High-performance model for agents and coding
39944002
39954003 - `string`
39964004
build-with-claude/claude-in-amazon-bedrock Changed · +4 / -4 lines
from line 102
102102 <Tabs>
103103 <Tab title="Gradle">
104104 ```kotlin
105 implementation("com.anthropic:anthropic-java:2.66.0")
106 implementation("com.anthropic:anthropic-java-bedrock:2.66.0")
105 implementation("com.anthropic:anthropic-java:2.67.0")
106 implementation("com.anthropic:anthropic-java-bedrock:2.67.0")
107107 ```
108108 </Tab>
109109
from line 112
112112 <dependency>
113113 <groupId>com.anthropic</groupId>
114114 <artifactId>anthropic-java</artifactId>
115 <version>2.66.0</version>
115 <version>2.67.0</version>
116116 </dependency>
117117 <dependency>
118118 <groupId>com.anthropic</groupId>
119119 <artifactId>anthropic-java-bedrock</artifactId>
120 <version>2.66.0</version>
120 <version>2.67.0</version>
121121 </dependency>
122122 ```
123123 </Tab>
build-with-claude/claude-in-microsoft-foundry Changed · +4 / -4 lines
from line 80
8080 <Tabs>
8181 <Tab title="Gradle">
8282 ```kotlin
83 implementation("com.anthropic:anthropic-java:2.66.0")
84 implementation("com.anthropic:anthropic-java-foundry:2.66.0")
83 implementation("com.anthropic:anthropic-java:2.67.0")
84 implementation("com.anthropic:anthropic-java-foundry:2.67.0")
8585
8686 // For Entra ID authentication, also add the Azure Identity library
8787 implementation("com.azure:azure-identity:1.18.3")
from line 93
9393 <dependency>
9494 <groupId>com.anthropic</groupId>
9595 <artifactId>anthropic-java</artifactId>
96 <version>2.66.0</version>
96 <version>2.67.0</version>
9797 </dependency>
9898 <dependency>
9999 <groupId>com.anthropic</groupId>
100100 <artifactId>anthropic-java-foundry</artifactId>
101 <version>2.66.0</version>
101 <version>2.67.0</version>
102102 </dependency>
103103 <!-- For Entra ID authentication, also add the Azure Identity library -->
104104 <dependency>
build-with-claude/claude-on-amazon-bedrock-legacy Changed · +4 / -4 lines
from line 54
5454 <Tab title="Java">
5555 <CodeGroup>
5656 ```groovy Gradle
57 implementation("com.anthropic:anthropic-java:2.66.0")
58 implementation("com.anthropic:anthropic-java-bedrock:2.66.0")
57 implementation("com.anthropic:anthropic-java:2.67.0")
58 implementation("com.anthropic:anthropic-java-bedrock:2.67.0")
5959 ```
6060
6161 ```xml Maven
from line 62
6262 <dependency>
6363 <groupId>com.anthropic</groupId>
6464 <artifactId>anthropic-java</artifactId>
65 <version>2.66.0</version>
65 <version>2.67.0</version>
6666 </dependency>
6767 <dependency>
6868 <groupId>com.anthropic</groupId>
6969 <artifactId>anthropic-java-bedrock</artifactId>
70 <version>2.66.0</version>
70 <version>2.67.0</version>
7171 </dependency>
7272 ```
7373
build-with-claude/claude-on-vertex-ai Changed · +4 / -4 lines
from line 45
4545 <Tab title="Java">
4646 <CodeGroup exclude="shell, python, typescript, csharp, go, php, ruby">
4747 ```groovy Gradle
48 implementation("com.anthropic:anthropic-java:2.66.0")
49 implementation("com.anthropic:anthropic-java-vertex:2.66.0")
48 implementation("com.anthropic:anthropic-java:2.67.0")
49 implementation("com.anthropic:anthropic-java-vertex:2.67.0")
5050 ```
5151
5252 ```xml Maven
from line 53
5353 <dependency>
5454 <groupId>com.anthropic</groupId>
5555 <artifactId>anthropic-java</artifactId>
56 <version>2.66.0</version>
56 <version>2.67.0</version>
5757 </dependency>
5858 <dependency>
5959 <groupId>com.anthropic</groupId>
6060 <artifactId>anthropic-java-vertex</artifactId>
61 <version>2.66.0</version>
61 <version>2.67.0</version>
6262 </dependency>
6363 ```
6464
build-with-claude/claude-platform-on-aws Changed · +4 / -4 lines
from line 305
305305
306306 <Tab title="Java">
307307 ```kotlin Gradle
308 implementation("com.anthropic:anthropic-java:2.66.0")
309 implementation("com.anthropic:anthropic-java-aws:2.66.0")
308 implementation("com.anthropic:anthropic-java:2.67.0")
309 implementation("com.anthropic:anthropic-java-aws:2.67.0")
310310 ```
311311
312312 ```xml Maven
from line 313
313313 <dependency>
314314 <groupId>com.anthropic</groupId>
315315 <artifactId>anthropic-java</artifactId>
316 <version>2.66.0</version>
316 <version>2.67.0</version>
317317 </dependency>
318318 <dependency>
319319 <groupId>com.anthropic</groupId>
320320 <artifactId>anthropic-java-aws</artifactId>
321 <version>2.66.0</version>
321 <version>2.67.0</version>
322322 </dependency>
323323 ```
324324 </Tab>
build-with-claude/preserved-thinking Changed · +3 / -3 lines
from line 36
3636
3737Claude Fable 5.1 and Claude Mythos 5.1 read thinking blocks produced by each other and by earlier Claude models. No earlier model reads thinking blocks from Claude Fable 5.1 or Claude Mythos 5.1.
3838
39Claude Opus 5.5 reads thinking blocks from Claude Opus 5, from earlier Opus, Sonnet, and Haiku models, and, on the Claude API, from Claude Sonnet 5.5, but not from Claude Fable or Claude Mythos models. On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read thinking blocks from Claude Opus 5.5; no other model does. So a conversation that moves from Claude Opus 5 onto Claude Opus 5.5 keeps its reasoning, and so does one that moves from Claude Opus 5.5 up to Claude Fable 5.1 or Claude Mythos 5.1 on the Claude API. One that moves from Claude Fable 5.1 or Claude Mythos 5.1 to Claude Opus 5.5, or from Claude Opus 5.5 to any model other than those two, runs the turns after the switch without the previous model's reasoning. The blocks are dropped, not rejected, as described below.
39Claude Opus 5.5 reads thinking blocks from Claude Opus 5, from earlier Opus, Sonnet, and Haiku models, and, on the Claude API and Google Cloud, from Claude Sonnet 5.5, but not from Claude Fable or Claude Mythos models. On the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read thinking blocks from Claude Opus 5.5; no other model does. So a conversation that moves from Claude Opus 5 onto Claude Opus 5.5 keeps its reasoning, and so does one that moves from Claude Opus 5.5 up to Claude Fable 5.1 or Claude Mythos 5.1 on the Claude API. One that moves from Claude Fable 5.1 or Claude Mythos 5.1 to Claude Opus 5.5, or from Claude Opus 5.5 to any model other than those two, runs the turns after the switch without the previous model's reasoning. The blocks are dropped, not rejected, as described below.
4040
41Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet 5, Claude Opus 4.8, Claude Haiku 4.5, and earlier models, but not from Claude Opus 5, Claude Opus 5.5, or any Claude Fable or Claude Mythos model. On the Claude API, Claude Opus 5.5 reads thinking blocks from Claude Sonnet 5.5; no other model does. So a conversation that moves from Claude Sonnet 5 onto Claude Sonnet 5.5 keeps its reasoning, and so does one that moves from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API. One that moves onto Claude Sonnet 5.5 from Claude Opus 5, Claude Opus 5.5, or a Claude Fable or Claude Mythos model runs the turns after the switch without the previous model's reasoning. So does any other move away from Claude Sonnet 5.5, for example a [server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback) to Claude Sonnet 5.
41Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet 5, Claude Opus 4.8, Claude Haiku 4.5, and earlier models, but not from Claude Opus 5, Claude Opus 5.5, or any Claude Fable or Claude Mythos model. On the Claude API and Google Cloud, Claude Opus 5.5 reads thinking blocks from Claude Sonnet 5.5; no other model does. So a conversation that moves from Claude Sonnet 5 onto Claude Sonnet 5.5 keeps its reasoning, and so does one that moves from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API and Google Cloud. One that moves onto Claude Sonnet 5.5 from Claude Opus 5, Claude Opus 5.5, or a Claude Fable or Claude Mythos model runs the turns after the switch without the previous model's reasoning. So does any other move away from Claude Sonnet 5.5, for example a [server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback) to Claude Sonnet 5.
4242
4343* **A conversation that moves to Claude Fable 5.1 from an earlier model, or from Claude Opus 5.5 on the Claude API, keeps its reasoning.** The earlier model's thinking blocks stay readable, so the model thinks as usual from the first turn after the switch.
4444* **A conversation that moves down to an earlier model loses Claude Fable 5.1's reasoning for that request.** This happens when a router sends a turn to a cheaper model, after a [classifier refusal fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback), or during a [server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback). The API removes the unreadable blocks before the prompt reaches the model. They aren't billed and don't count toward `input_tokens`.
build-with-claude/prompt-engineering/claude-prompting-best-practices Changed · +5 / -5 lines
from line 24
2424| Claude Fable 5 and Claude Mythos 5 | [Prompting Claude Fable 5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-fable-5) | Differences from Claude Opus 4.8: effort levels, instruction following, long-run progress claims, memory systems, and the `reasoning_extraction` refusal category. |
2525| Claude Sonnet 5.5 | [Prompting Claude Sonnet 5.5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-sonnet-5-5) | Differences from Claude Sonnet 5: effort calibration, initiative and scope, running without up-front thinking, JSON output on reasoning tasks, user-facing progress updates, tool use in chat, mid-turn user messages, verification on coding tasks, tool-call handling, complex visual inputs, and safeguard refusals. |
2626| Claude Sonnet 5 | [Prompting Claude Sonnet 5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-sonnet-5) | Differences from Claude Sonnet 4.6: response length, effort and thinking-depth calibration, tool use triggering, literal instruction following, and design and frontend defaults. |
27| Claude Opus 5.5 | [Prompting Claude Opus 5.5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5-5) | Differences from Claude Opus 5: effort calibration, prompts written for thinking disabled, user-facing progress updates, safeguard false positives, and tools for complex visual inputs. |
27| Claude Opus 5.5 | [Prompting Claude Opus 5.5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5-5) | Differences from Claude Opus 5: effort calibration, prompts written for thinking disabled, user-facing progress updates, safeguard refusals, and tools for complex visual inputs. |
2828| Claude Opus 5 | [Prompting Claude Opus 5](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5) | Differences from prior Opus models: response length and verbosity, user-facing progress updates, written deliverable length, task scope and over-verification, subagent control, and self-correction. |
2929| Claude Opus 4.8 | [Prompting Claude Opus 4.8](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-4-8) | Response length, effort and thinking-depth calibration, tool use triggering, literal instruction following, subagent control, and design and frontend defaults. |
3030
from line 327
327327* **More conversational:** Slightly more fluent and colloquial, less machine-like
328328* **Less verbose:** May skip detailed summaries for efficiency unless prompted otherwise
329329
330This means Claude may skip verbal summaries after tool calls, jumping directly to the next action. If you prefer more visibility into its reasoning:
330This means Claude may skip verbal summaries after tool calls, jumping directly to the next action. If you prefer more visibility into its work:
331331
332332```text Sample prompt wrap
333333After completing a task that involves tool use, provide a quick summary of the work you've done.
from line 776
776776If you are not using extended thinking, no changes are required. On Claude Opus 4.6 through Claude Opus 4.8 and Claude Sonnet 4.6, thinking is off when you omit the `thinking` parameter. On Claude Opus 5, Claude Sonnet 5.5, and Claude Sonnet 5, thinking is on by default when you omit the `thinking` parameter. On Claude Opus 5, you can disable it only at effort `high` or lower. On Claude Sonnet 5.5, the lowest thinking setting is `between_tools`, accepted at effort `high` or lower. On Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, and Claude Opus 5.5, thinking is always on, regardless of whether you set the `thinking` parameter.
777777
778778* **Prefer general instructions over prescriptive steps.** A prompt like "think thoroughly" often produces better reasoning than a hand-written step-by-step plan. Claude's reasoning frequently exceeds what a human would prescribe.
779* **Multishot examples work with thinking.** Use `<thinking>` tags inside your few-shot examples to show Claude the reasoning pattern. It will generalize that style to its own extended thinking blocks.
780* **Manual chain-of-thought (CoT) prompting as a fallback.** When thinking is off, you can still encourage step-by-step reasoning by asking Claude to think through the problem. Use structured tags like `<thinking>` and `<answer>` to cleanly separate reasoning from the final output. On Claude Opus 5, prefer keeping thinking enabled at a lower effort level instead: with thinking disabled, the model can occasionally emit internal XML tags into its visible output, so see [Running with thinking disabled](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5#running-with-thinking-disabled) before applying this pattern there.
779* **Multishot examples work with thinking.** Worked examples in your prompt shape how Claude approaches similar problems in its own thinking blocks. Present each example as a problem, the method to apply, and the expected answer.
780* **Manual chain-of-thought (CoT) prompting as a fallback.** When thinking is off, you can still encourage step-by-step reasoning by asking Claude to think through the problem before it answers, and to put the final answer in `<answer>` tags so you can extract it. On Claude Fable 5.1, Claude Fable 5, Claude Opus 5.5, Claude Opus 5, and Claude Sonnet 5.5, rely on thinking instead, at a lower [effort](https://platform.claude.com/docs/en/build-with-claude/effort) level if cost or latency matters: a prompt that asks the model to write out its reasoning, for example in `<thinking>` tags, may be declined. See [Keep reasoning in thinking blocks](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#keep-reasoning-in-thinking-blocks).
781781* **Ask Claude to self-check.** Append something like "Before you finish, verify your answer against \[test criteria]." This catches errors reliably, especially for coding and math. Claude Opus 5 is the exception: it verifies its own work well without explicit instruction, and verification instructions carried over from prompts tuned for earlier models can cause over-verification, adding tokens and latency. On Claude Opus 5, remove these instructions rather than rewriting them. See [Task scope and over-verification](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5#task-scope-and-over-verification).
782782
783783<Note>
from line 1110
11101110 </Card>
11111111
11121112 <Card title="Prompting Claude Opus 5.5" icon="terminal" href="https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5-5">
1113 Behavioral differences and prompting patterns for Claude Opus 5.5, covering effort calibration, thinking always on, progress updates, safeguard false positives, and visual inputs.
1113 Behavioral differences and prompting patterns for Claude Opus 5.5, covering effort calibration, thinking always on, progress updates, safeguard refusals, and visual inputs.
11141114 </Card>
11151115
11161116 <Card title="Prompting Claude Opus 5" icon="terminal" href="https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5">
build-with-claude/prompt-engineering/prompting-claude-opus-5 Changed · +4 / -0 lines
## Reasoning in the response
from line 102
102102
103103Instructions that call out thinking tags by name are less effective than the general form, so avoid naming them specifically.
104104
105## Reasoning in the response
106
107Prompts, skills, and tool descriptions that ask Claude Opus 5 to write out its thinking or reasoning, verbatim or in a fixed format, may be declined with the `reasoning_extraction` [refusal category](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#refusal-response). Ask for a short explanation of the answer or a summary of the actions taken instead, and, if you need the reasoning, keep thinking on and read the [summarized thinking](https://platform.claude.com/docs/en/build-with-claude/thinking#summarized-thinking) blocks (`display: "summarized"`). See [Keep reasoning in thinking blocks](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#keep-reasoning-in-thinking-blocks).
108
build-with-claude/refusals-and-fallback Changed · +22 / -9 lines
### Keep reasoning in thinking blocks
from line 180
180180* `category` and `explanation` are both `null` when the refusal does not map to a named category. That `null` is a normal, permanent value, not a placeholder.
181181* `stop_details` itself is `null` for every stop reason other than `refusal`.
182182
183| `category` | What it means | Billed before any output |
184| ------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------ |
185| `"cyber"` | The request could enable cyber harm, such as malware or exploit development. Benign cybersecurity work can also trigger this category. | No |
186| `"bio"` | The request could enable biological harm, such as dangerous lab methods. Beneficial life sciences work can also trigger this category. | Yes |
187| `"frontier_llm"` | The request could assist the development of competing AI models, which is restricted under [Anthropic's commercial terms](https://www.anthropic.com/legal/commercial-terms). Benign machine learning work can also trigger this category. | Yes |
188| `"reasoning_extraction"` | The request asks the model to reproduce its internal reasoning in the response text. To get reasoning in a structured form instead, use [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking). | Yes |
189| `"general_harms"` | The request falls under a usage-policy area outside the four named categories. Benign work can also trigger this category. | No |
183| `category` | What it means | Billed before any output |
184| ------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ------------------------ |
185| `"cyber"` | The request could enable cyber harm, such as malware or exploit development. Benign cybersecurity work can also trigger this category. | No |
186| `"bio"` | The request could enable biological harm, such as dangerous lab methods. Beneficial life sciences work can also trigger this category. | Yes |
187| `"frontier_llm"` | The request could assist the development of competing AI models, which is restricted under [Anthropic's commercial terms](https://www.anthropic.com/legal/commercial-terms). Benign machine learning work can also trigger this category. | Yes |
188| `"reasoning_extraction"` | The request asks the model to reproduce its internal reasoning in the response text. To get reasoning in a structured form instead, use [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking). See [Keep reasoning in thinking blocks](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#keep-reasoning-in-thinking-blocks). | Yes |
189| `"general_harms"` | The request falls under a usage-policy area outside the four named categories. Benign work can also trigger this category. | No |
190190
191191A refusal can arrive before any output, or mid-stream after partial output. In either case, treat any partial output as incomplete and discard it.
192192
193### Keep reasoning in thinking blocks
194
195A `reasoning_extraction` refusal usually comes from a prompt that asks the model to put its thinking or reasoning in the output, either verbatim or in a fixed format. Common examples:
196
197* A `<thinking>`, `<reasoning>`, or scratchpad section that the model fills in before it answers
198* A `reasoning`, `thinking`, or `trace` field in JSON output or in a tool input
199* The model's private notes, or a running log of its reasoning
200* Reasoning asked for verbatim or in full
201
202The wording can sit in a system prompt, a skill, or a tool description, so check those too.
203
204You can still ask Claude to explain its answer. Ask for a short explanation, the evidence behind a result, or a summary of the actions it took. To see how the model reasoned, set `display: "summarized"` and read the [`thinking` blocks](https://platform.claude.com/docs/en/build-with-claude/thinking#summarized-thinking). This category has no recommended fallback model, so change the prompt rather than retry the request. If a reworded request is still declined, [contact support](https://support.claude.com/en/articles/9015913-how-to-get-support) and include the [request ID](https://platform.claude.com/docs/en/api/errors#request-id).
205
193206## How refusals are billed
194207
195208These billing rules apply on every platform: the Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, and Microsoft Foundry.
from line 1128
11151128
11161129* Retries walk your fallback list in order. A fallback model that itself refuses passes the request to the next entry.
11171130* When every model in the list has declined, the middleware returns the final refusal (the last model's refusal response) rather than raising an error.
1118* Thinking blocks from Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, or Claude Fable 5 pass through unchanged. Each retry re-sends your original request body, and the only blocks the middleware removes from conversation history on later requests are the `fallback` boundary blocks it added itself. The fallback model can't read Claude Fable 5.1 blocks, which are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them. The API also drops Claude Opus 5.5 blocks for every fallback model except Claude Fable 5.1 and Claude Mythos 5.1 (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API.
1131* Thinking blocks from Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, or Claude Fable 5 pass through unchanged. Each retry re-sends your original request body, and the only blocks the middleware removes from conversation history on later requests are the `fallback` boundary blocks it added itself. The fallback model can't read Claude Fable 5.1 blocks, which are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them. The API also drops Claude Opus 5.5 blocks for every fallback model except Claude Fable 5.1 and Claude Mythos 5.1 (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API and Google Cloud.
11191132* Responses served through the middleware include a `fallback` content block at each model boundary, the same as server-side fallback responses. The middleware manages those blocks for you on later requests.
11201133* The model that accepted is recorded in `BetaFallbackState`, so follow-up requests that share the state stay pinned to it rather than re-asking a model that refused.
11211134
from line 1148
11351148 <Step title="Re-send on a fallback model">
11361149 Send the same request with `model` set to a fallback model, such as Claude Opus 4.8. If the refused request sent `thinking: {"type": "between_tools"}`, change `thinking` first: only Claude Sonnet 5.5 accepts that value, so omit `thinking` or set a value the fallback model accepts. [Server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback) under the `2026-07-01` header makes this change for you when it falls back to Claude Sonnet 5. Another model can normally serve a request that Claude Fable 5.1 or Claude Fable 5 declines. How you handle the conversation history depends on whether you redeem a [fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit):
11371150
1138 * **Not redeeming a credit:** you can leave the earlier `thinking` and `redacted_thinking` blocks in place or strip them to save input tokens. The fallback model normally can't use them either way: it ignores Claude Fable 5 blocks, and Claude Fable 5.1 blocks are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them. The API also drops Claude Opus 5.5 blocks for every fallback model except Claude Fable 5.1 and Claude Mythos 5.1 (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API.
1151 * **Not redeeming a credit:** you can leave the earlier `thinking` and `redacted_thinking` blocks in place or strip them to save input tokens. The fallback model normally can't use them either way: it ignores Claude Fable 5 blocks, and Claude Fable 5.1 blocks are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them. The API also drops Claude Opus 5.5 blocks for every fallback model except Claude Fable 5.1 and Claude Mythos 5.1 (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API and Google Cloud.
11391152 * **Redeeming a credit:** send the body unchanged, because redemption requires an exact match. The server handles the earlier model's thinking blocks on a redemption, so do not strip them (see [Fields that must match the refused request](https://platform.claude.com/docs/en/build-with-claude/fallback-credit#reference)).
11401153 </Step>
11411154
claude_api_primer Changed · +13 / -21 lines
### Chain of thought
from line 633
633633
634634Tools do not necessarily need to be client functions. You can use tools anytime you want the model to return JSON output that follows a provided schema.
635635
636### Chain of thought
636### Parallel tool use
637637
638When using tools, Claude often shows its "chain of thought," that is, the step-by-step reasoning it uses to break down the problem and determine which tools to use.
638By default, Claude may use multiple tools to answer a user query. You can disable this behavior by setting `disable_parallel_tool_use=true`.
639639
640## Handling tool use and tool result content blocks
641
642### Handling results from client tools
643
644The response has a `stop_reason` of `tool_use` and one or more `tool_use` content blocks that include:
645
646* `id`: A unique identifier for this particular tool use block.
647* `name`: The name of the tool being used.
648* `input`: An object containing the input being passed to the tool.
649
640650```json
641651{
642652 "role": "assistant",
643653 "content": [
644654 {
645 "type": "text",
646 "text": "<thinking>To answer this question, I will: 1. Use the get_weather tool to get the current weather in San Francisco. 2. Use the get_time tool to get the current time in the America/Los_Angeles timezone, which covers San Francisco, CA.</thinking>"
647 },
648 {
649655 "type": "tool_use",
650656 "id": "toolu_01A09q90qw90lq917835lq9",
651657 "name": "get_weather",
from line 660
654660 ]
655661}
656662```
657
658### Parallel tool use
659
660By default, Claude may use multiple tools to answer a user query. You can disable this behavior by setting `disable_parallel_tool_use=true`.
661
662## Handling tool use and tool result content blocks
663
664### Handling results from client tools
665
666The response has a `stop_reason` of `tool_use` and one or more `tool_use` content blocks that include:
667
668* `id`: A unique identifier for this particular tool use block.
669* `name`: The name of the tool being used.
670* `input`: An object containing the input being passed to the tool.
671663
672664When you receive a tool use response, you should:
673665