Tasks and inputs
Claude Mythos 5.2 is listed for Text generation, Coding & Code Generation, Computer Use. It accepts text, image and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Nov 17, 2026
Claude Mythos 5.2 is the verified-access edition of Claude Fable 5.2 for vetted cyber and life-sciences organisations, with a 1M context at $8/$40.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Pricing conditions: Indicative: $8 input / $40 output per 1M tokens; cache reads $0.20; cache writes $10 (5-min) / $16 (1-h); Batch 50% off; fast mode $16 / $80. Restricted access; 30-day data retention required. Confirm with your Anthropic account team.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
What you apply for here is access, not a different model. Anthropic describes Claude Mythos 5.2 as the same model as Claude Fable 5.2, made available only to organisations verified through its verification programmes. Both shipped on 9 November 2026. The specification sheet, the price and the breaking changes are identical; the gate, and the work the gate permits, is what differs.
The programmes named for access are the Cyber Verification Program and a Life Sciences programme. To request access, an organisation applies to whichever covers its use case, or goes through its account team. Nothing on this page can be tried before that clears, so security and research leads should start the application before they start the evaluation plan.
The restricted line is not new. It follows Claude Mythos 5 and Mythos 5.1, each of which mirrored its Fable sibling.
| Field | Value |
|---|---|
| API ID | claude-mythos-5-2 |
| Released | 9 November 2026 |
| Context window | 1M tokens |
| Max output | 128K tokens |
| Input / output | Text and image in, text out |
| Thinking | Adaptive, always on; default effort high |
| Comparative latency | Moderate |
| Knowledge cutoff | June 2026 |
| Retirement | Not before 9 November 2027 |
| Base price | $8 input / $40 output per 1M tokens |
Base rates are $8 and $40 per million tokens. Cache reads cost $0.20, five-minute cache writes $10 and one-hour writes $16. Batch halves the base. Fast mode is $16 and $80, available on the Claude API and on other platforms Anthropic lists separately.
Some launch coverage gave Fable's price but none for Mythos. Anthropic itself lists $8 and $40 for both. Anthropic's figure is the one to use, but verified-access agreements can carry their own terms, so confirm with your account team. Every figure on this page is indicative.
claude-mythos-5-2global.anthropic.claude-mythos-5-2 or us.anthropic.claude-mythos-5-2, bedrock-runtime onlyLike Fable 5.2, it requires 30-day data retention and is unavailable under zero data retention unless Anthropic expressly authorises it.
The upgrade is a 20% price cut, from $10 and $50, with output reported up to 30% faster. The cost is a migration. Prefix checks now fail loudly with a 400 when code edits earlier turns and replays a thinking block. The computer_20251124 tool is rejected outside Bedrock in favour of computer_toolset_20260801. Fable 5.2 and Mythos 5.2 read each other's thinking blocks, so switching between the two keeps reasoning intact, but no other model reads theirs.
The advisor tool deserves a line of its own. When Mythos 5.2 or Fable 5.2 is the executor, only those two are accepted as advisors.
Approved security teams should replay a closed incident: the logs, the vulnerable code and the eventual fix. Run it at high and again at xhigh, with the same tools your analysts used. Compare the findings, the token bill and the time saved. If the gain only appears at xhigh, price it at that effort level.
For work that does not need verified access, Claude Fable 5.1 and Claude Opus 5.5 are open to all API customers, and the full family is on the Anthropic provider page.
Compare the model on the work you need to complete.
Claude Mythos 5.2 is listed for Text generation, Coding & Code Generation, Computer Use. It accepts text, image and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.
Documented limits: 1M context and 128K maximum output. The listed route has starting rates of $8.00 input and $40.00 output per million tokens. Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.
Use the model identifier and account requirements for the selected API route. Confirm its request limits and output format with one complete task before connecting a larger workflow.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
Tencent
Tencent Hy4 Flash is the cost-focused companion to Hy4, keeping vision and agent skills at a far lower token price. Availability is still unsettled.
Alibaba
Qwen4 Max (qwen4-max) is Alibaba's first Qwen4-generation flagship: 1M-token context, text, image and video input, thinking on by default, $1.50/$4.50 per 1M.
Sarvam
Sarvam-105B-1711 is a refreshed open post-train of Sarvam's 105B MoE, tuned for tool calling, instruction following, agents and Indian languages.
Alibaba
Qwen4 Max Preview was Alibaba's five-day early-access build of its Qwen4 flagship, offered only through a Token Plan subscription before GA qwen4-max.
Gemini 4 Carbon is Google's strongest Gemini 4 reasoning model for long-horizon agentic coding, scientific research and cyber defence, with a 1M context.
Anthropic
Claude Fable 5.2 is Anthropic's top model for demanding reasoning and long-horizon agentic work, with a 1M context at $8/$40, 20% below Fable 5.1.