Follow Discord
Sweep 22 Oct 2026 · 04:32Z Build v2.1.308 527 read Stable v2.1.308 Latest v2.1.308 Next v2.1.308 Feeds RSS JSON llms.txt llms-full.txt Unofficial
HomeReleasesv2.1.283entry #86

Claude Code v2.1.283 ·

Opus [1m] suffix and "with 1M context" label now depend on each model's 1M support

Default Opus, Opus subagents and the picker label add 1M context only where a model's 1M window is opt-in, not native

Group of 3 You'll notice Improvements
JSON All of v2.1.283
You'll noticeTier: how much it should matter to you
3Useful: my rating, 1 to 5
3Signal: worth watching, 1 to 5
ModelsArea: what it touches
ImprovementsKind: in v2.1.283,
ImprovementsSection of the release

Unclear It is not clear which Opus models offer one-million-token context by default, as an option, or never.

What

The context window is how much text a model can consider at once. Some models have a 1M-token window built in (native). Others offer it only when asked for (opt-in), which Claude Code does by adding [1m] to the model name. Claude Code now tracks each model's 1M support as "never", "native" or "optIn" instead of a yes or no:

  • CLAUDE_CODE_DISABLE_1M_CONTEXT makes every model count as "never".
  • Default Opus: [1m] is added to the default Opus model only when its support is "optIn". Models with a native 1M window no longer get [1m].
  • Picking opus: when the opus alias points to a model with opt-in 1M, choosing it now resolves to that model with [1m]. The option label also shows the resolved model id.
  • Subagents: an Opus-family subagent model gets [1m] when it lacks it and its 1M support is "optIn" on a first-party provider.
  • Model picker: the Opus description shows "with 1M context" based on the selected model's support being "optIn", instead of one global check. Opus counts as 1M-eligible unless its support is "never".

Why

The model name shown for default Opus may change, and Opus subagents may now run with a larger context window, which can affect cost. The picker should stop claiming 1M context for Opus models that do not offer it.

Read from
Names in the bundleANTHROPIC_DEFAULT_OPUS_MODEL
Since it was published

The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.

Added since A small documentation edit on Model configuration touched a line naming ANTHROPIC_DEFAULT_OPUS_MODEL after this was published. * **[Auto mode classifier](/docs/en/permission-modes#eliminate-prompts-with-auto-mode)**: the classifier's Claude Sonnet 5 default applies only when the allowlist permits Sonnet 5. When it's excluded, the classifier runs on the session's m… model-config see the edit
Confirmed since Anthropic's documentation has since written up ANTHROPIC_DEFAULT_OPUS_MODEL, on Model configuration. * An Opus target must resolve in your deployment, whichever model refused: set `ANTHROPIC_DEFAULT_OPUS_MODEL`, or keep an Opus 4.8 entry in the provider's model list. Without one, fallback stays off for every source model, including Sonnet… model-config see the edit
How sure we are
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtIt is not clear which Opus models offer one-million-token context by default, as an option, or never.
Anthropic's documentation agreesAnthropic's documentation has since written up ANTHROPIC_DEFAULT_OPUS_MODEL, on Model configuration.

See this entry in the whole of v2.1.283 →

Feedback