optimizing-for-cost-and-intelligence changedabout-claude/models/optimizing-for-cost-and-intelligence
Nearest release: v2.1.308, published 2 hours after this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+5added
Lines−4removed
From line
61
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits24to this page, all time
The whole hunk
from line 61, old and new numbered
/
from line 61
6161* Deep in a task, a well-built loop pays full price on under 1% of its input.
6262* Below about 80%, look for something breaking the cache (see [What breaks the cache](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#what-breaks-the-cache)).
6363
64Across Anthropic's measured runs, cache reads are routinely the largest single component of task cost, making caching worth more than most model-choice decisions. Anthropic priced the DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) runs with and without caching:
64Anthropic priced the DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) runs with and without caching:
6565
6666<Frame>
6767 
from line 319
319319
320320* Each model generation responds to prompts differently, so a prompt accumulates text written for a model you no longer use.
321321* The usual case is over-specific instruction added to compensate for an older model: "verify twice," "be maximally thorough," a mandatory step-by-step procedure, or a hand-rolled reasoning scratchpad.
322* A newer model follows such instructions to the letter, producing extra tool rounds and extra writing, so the bill goes up with no gain in accuracy.
322* A newer model follows such instructions to the letter, producing extra tool rounds and extra writing that you pay for with no gain in accuracy.
323323
324324The audit is one command. The [Claude API skill](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/claude-api-skill) that ships with Claude Code has a `prompt-audit` command that reads a project's prompts and request code and reports what was written for a different model:
325325
from line 424
424424
425425### Upgrade the model
426426
427If you are a model or two behind, the cheapest lever is the model string. In Anthropic's measurements, each newer model solved at least as many tasks as the one before it, usually for less per solved task. The chart plots recent Claude models on the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset, where every model ran through the same test setup, and on Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs); the table lists what each upgrade changed:
427If you are a model or two behind, the cheapest lever is the model string. In Anthropic's measurements, each newer model solved at least as many tasks as the one before it, usually for less per solved task. The chart plots recent Claude models on two benchmarks in which an agent works alone on one written task and tests check the result: the SWE-bench Pro[3](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs) subset, where every model ran through the same test setup, and Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs). The table lists what each upgrade changed:
428428
429429<Frame>
430430 
from line 444
444444
445445**When it pays.**
446446
447* The size of the saving depends on the work. From Claude Opus 5 to Claude Opus 5.5, it is 78% per solved task on the SWE-bench Pro subset and 40% on typical workloads ([Introducing Claude Opus 5.5](https://www.anthropic.com/news/claude-opus-5-5)). From Claude Sonnet 5 to Claude Sonnet 5.5, it is 87% on the same subset and up to 30% for most work ([Introducing Claude Sonnet 5.5](https://www.anthropic.com/news/claude-sonnet-5-5)).
447448* Where older models already solve most tasks, the cost of each attempt decides the saving: on the SWE-bench Pro subset, Claude Opus 5.5 scores about the same as Claude Opus 5 for 78% less per solved task.
448449* Where older models mostly fail, the pass rate decides the saving: on Terminal-Bench 3[20](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), the upgrade from Claude Opus 4.8 to Claude Opus 5 raises tasks solved from 15% to 41% and costs 56% less per solved task.
449450* Measure the upgrade on your own workload before assuming it saves. The saving is not guaranteed: on DeepResearch Bench II[7](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#refs), Claude Fable 5.1 costs more per task than Claude Fable 5 because the new model does more work per task there, although the input and output prices are the same and the cache read is 4x cheaper.
from line 862
861862| Data files through code execution | 92% on a data task of 25 questions, with prompt caching off | A gain: 25 of 25 correct instead of 7 | Faster | [Trim input and context tokens](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#trim-input-and-context-tokens) |
862863| Batch API | 50% | None | Results within 24 hours | [Batch work that can wait](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#batch-work-that-can-wait) |
863864| Prompt audit against the current model | 14% on both model upgrades tested | None; a gain on one | Faster (fewer tool rounds) | [Audit prompts against the current model](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#audit-prompts-against-the-current-model) |
864| Upgrade the model | Usually less per solved task: up to 87% less on a coding benchmark | None; a gain on most | Neutral | [Upgrade the model](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#upgrade-the-model) |
865| Upgrade the model | Usually less: up to 40% less on typical workloads, and up to 87% less per solved task on a coding benchmark | None; a gain on most | Neutral | [Upgrade the model](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#upgrade-the-model) |
865866| Lower effort | On long coding tasks, `medium` cost about 30% to 52% less than `high`, and `low` about two thirds less | Up to 15 points lower, depending on the task; none on one research agent | Faster | [Tune effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#tune-effort) |
866867| Re-run failures | About 40% less than running everything at `high`, at the same pass rate or slightly better | None | A second attempt on the tasks that fail | [Re-run failures at higher effort](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#re-run-failures-at-higher-effort) |
867868| Ask for shorter answers | About 13% less, from 39% fewer output tokens | None | Faster | [Set budgets and output caps](https://platform.claude.com/docs/en/about-claude/models/optimizing-for-cost-and-intelligence#set-budgets-and-output-caps) |
No line in this hunk matches that.