Data as of November 10, 2026 · How the score is built
Released Nov 9, 20261M contextClaude Fable 5.2 and Claude Mythos 5.2 system card
Claude Mythos 5.2
Decision readingClaude Mythos 5.2 is tracked, but not publicly ranked yet. The profile exposes 16 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Released Nov 9, 2026 — see all recent releases
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
Unranked
field median 51.2Not eligible for a public rank
Price
$8input / $40 output
input median $0.60cached $0.20 · blended $24
Speed
Not measured
field median 92 tok/sTime to first token not measured
Context
1Mtokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Published rows are visible, but no category has enough eligible evidence for a comparative rank.
Validate before choosing
16 published rows leave some tracked benchmark slots empty. Independent runtime speed has not been measured.
Source-linked · 16 displayable benchmark rows
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%5 benchmarksVerified | Score pending | 5 benchmarks | Verified | |||
| CodingWeight 20%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%1 benchmarkVerified | Score pending | 1 benchmark | Verified | |||
| KnowledgeWeight 12%8 benchmarksVerified | Score pending | 8 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
16 of 890 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic5/5 verified
- CodingNot measured
- ReasoningNot measured
- Multimodal1/1 verified
- Knowledge8/8 verified
- MultilingualNot measured
- Inst. FollowingNot measured
- MathNot measured
Benchmark ledger
Agentic opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Agentic5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 4.0 | Score70.50% | Versus best verified row Best verified: Claude Sonnet 5.5 · 70.60% | Gap0.1 behind | Weight8% ref. weight | |
| Terminal-Bench-Science 0.1 | Score75.3% | Versus best verified row Best verified: Claude Mythos 5.2 · 75.3% | GapBest verified | WeightDisplay only | |
| CVE RediscoveryAnthropic CVE Rediscovery | Score89.6% | Versus best verified row Best verified: Claude Mythos 5.2 · 89.6% | GapBest verified | WeightDisplay only | |
| ExploitGym | Score38.1% | Versus best verified row Best verified: Gemini 4 Carbon · 53.5% | Gap15.4 behind | WeightDisplay only | |
| Binary Exploitation BenchmarkAnthropic Binary Exploitation Benchmark (formerly OSS-Fuzz), control-flow hijacks | Score148 | Versus best verified row Best verified: Claude Mythos 5.2 · 148 | GapBest verified | WeightDisplay only |
Multimodal1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Biomedical image analysisAnthropic biomedical-image-analysis evaluation | Score80.7% | Versus best verified row Best verified: Claude Mythos 5.2 · 80.7% | GapBest verified | WeightDisplay only |
Knowledge8 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| De novo protein-binder designAnthropic de novo protein-binder design evaluation | Score85.1% | Versus best verified row Best verified: Claude Mythos 5.2 · 85.1% | GapBest verified | WeightDisplay only | |
| Protein DesignAnthropic Protein Design evaluation | Score64.6% | Versus best verified row Best verified: Claude Mythos 5.2 · 64.6% | GapBest verified | WeightDisplay only | |
| SingleCellBenchLatchBio SingleCellBench | Score65.6% | Versus best verified row Best verified: Claude Mythos 5.2 · 65.6% | GapBest verified | WeightDisplay only | |
| Protocols (troubleshooting)Molecular Biology Protocols Troubleshooting | Score77.1% | Versus best verified row Best verified: Claude Mythos 5.2 · 77.1% | GapBest verified | WeightDisplay only | |
| Protein Domain IDProtein domain identification | Score91.8% | Versus best verified row Best verified: Claude Mythos 5.2 · 91.8% | GapBest verified | WeightDisplay only | |
| Protein Mutation AttributionProtein mutation attribution | Score51.5% | Versus best verified row Best verified: Claude Mythos 5.2 · 51.5% | GapBest verified | WeightDisplay only | |
| Morphology-to-molecule matchingAxiom Bio morphology-to-molecule matching | Score35.2% | Versus best verified row Best verified: Claude Mythos 5.2 · 35.2% | GapBest verified | WeightDisplay only | |
| Medicinal chemistryAnthropic medicinal-chemistry evaluation | Score69.0% | Versus best verified row Best verified: Claude Mythos 5.2 · 69.0% | GapBest verified | WeightDisplay only |
External signals2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| CyScenarioBench successCyScenarioBench Average Success Rate | Score76% | Versus best verified row Best verified: Claude Mythos 5.2 · 76% | GapBest verified | WeightDisplay only | |
| ExploitBenchExploitBench v8-bench | Score95% | Versus best verified row Best verified: GPT-6 Astra · 100% | Gap5 behind | WeightDisplay only |
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- claude-mythos-5-2Anthropic Claude Mythos 5.2 model documentation
- Context window
- 1MAnthropic Claude Mythos 5.2 model documentation
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, imageAnthropic Claude Mythos 5.2 model documentation
- Output modalities
- textAnthropic Claude Mythos 5.2 model documentation
- Parameters
- Not disclosed by the provider
- Availability
- Cyber Verification Program · Life Sciences Verification Program · Claude SecurityAnthropic Claude Mythos 5.2 model documentation
- Cloud regions
- Not tracked yet
- Lifecycle
- limited accessAnthropic Claude Mythos 5.2 model documentation
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.20 per million cached input tokensClaude Mythos 5.2 model documentation
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Claude Mythos 5.2, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.
Claude Mythos 5.2 is a proprietary model with a 1M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Limited to organizations verified through Anthropic's Cyber Verification Program and Life Sciences Verification Program, as `claude-mythos-5-2` on the Claude API, Claude Models on Google Cloud, and Microsoft Foundry. Claude Security, available to all Claude Enterprise customers, runs on Mythos 5.2.
Anthropic says Claude Mythos 5.2 and Claude Fable 5.2 are the same underlying model; Mythos 5.2 runs without certain cyber and biology safeguards. We do not copy Fable results onto the restricted profile and store only results Anthropic reports for Mythos 5.2 itself. Its rows come from the system card's safeguards-off runs: Terminal-Bench 4.0 and Terminal-Bench-Science, CVE Rediscovery, ExploitBench, CyScenarioBench, ExploitGym, and the life-science evaluations.
Claude Mythos 5.2 sits in the Claude Mythos family with Claude Mythos 5, Claude Mythos 5.1, Claude Mythos Preview. Its explicit predecessor is Claude Mythos 5.1. 16 of 890 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Last updated November 10, 2026. Runtime fields remain blank until a sourced snapshot exists.
Ask your AI, from current BenchLM data
Where does Claude Mythos 5.2 rank right now, and what does it cost?
Questions
How does Claude Mythos 5.2 perform overall in AI benchmarks?
Claude Mythos 5.2 has 16 source-displayable benchmark rows, but it does not qualify for a public overall rank. The available rows remain visible by category without being converted into a site-wide score. Missing evidence stays blank instead of being estimated from an earlier model.
Is Claude Mythos 5.2 good for knowledge and understanding?
Claude Mythos 5.2 has source-displayable benchmark coverage for knowledge and understanding, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Claude Mythos 5.2 good for agentic tool use and computer tasks?
Claude Mythos 5.2 has source-displayable benchmark coverage for agentic tool use and computer tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Claude Mythos 5.2 good for multimodal and grounded tasks?
Claude Mythos 5.2 has source-displayable benchmark coverage for multimodal and grounded tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Which sibling models are related to Claude Mythos 5.2?
Claude Mythos 5.2 belongs to the Claude Mythos family. Related tracked variants include Claude Mythos 5, Claude Mythos 5.1, Claude Mythos Preview. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Claude Mythos 5.2 have full benchmark coverage on BenchLM?
No. Claude Mythos 5.2 currently has 16 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Claude Mythos 5.2?
Claude Mythos 5.2 has a documented context window of 1M. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.