Skip to main content

Model profile

Claude Mythos 5

AnthropicCurrentReleased Jun 9, 2026
Data verified
Overall Score
83.93Public #1 of 200Verified #1 of 99
Arena Elo
Not listed
Eligible category ranks
3of 8
Price (1M tokens)
$10 in / $50 out
API pricing
Speed
Not listed
Context
1M+

Evidence coverage

15 of 323 tracked benchmarks are published. 15 are verified and 0 provisional. 6 of 8 categories are measured.

Updated July 23, 2026Methodology
Published / tracked
15 / 323
Verified
15
Provisional
0
Categories with evidence
6 / 8

Evidence by category

  • Agentic4 benchmarks
    Verified
  • Coding3 benchmarks
    Verified
  • Reasoning0 benchmarks
    Not measured
  • Knowledge3 benchmarks
    Verified
  • Math1 benchmark
    Verified
  • Multilingual1 benchmark
    Verified
  • Multimodal3 benchmarks
    Verified
  • Inst. Following0 benchmarks
    Not measured
ProprietaryReasoning
Confidence:
Medium
restricted

Claude Mythos 5 ranks #1 out of 200 models on the public leaderboard with an overall score of 83.93/100. It also ranks #1 out of 99 on the verified leaderboard. This places it among the top tier of AI models available in 2026, competing directly with the strongest models from all major AI labs.

Claude Mythos 5 is a proprietary model with a 1M+ token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.

Claude Mythos 5 is restricted to Glasswing partners and trusted-access programs. Claude Fable 5 is the same underlying model made generally available with additional safeguards for cybersecurity, biology, chemistry, distillation, and frontier-AI-development requests.

This profile currently has 15 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Coding (#1), while its weakest is Agentic (#7). This performance profile makes it particularly well-suited for software development and code generation tasks.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 76.7283.93

  1. Claude Mythos 5Current model
    Anthropic
    #183.93
    Claude Mythos 5 is #1 with a score of 83.93.
  2. Claude Fable 5
    Anthropic
    #283.68
    Claude Fable 5 is #2 with a score of 83.68.
    Compare
  3. GPT-5.6 Sol
    OpenAI
    #381.96
    GPT-5.6 Sol is #3 with a score of 81.96.
    Compare
  4. Kimi K3
    Moonshot AI
    #480.96
    Kimi K3 is #4 with a score of 80.96.
    Compare
  5. Claude Opus 4.8
    Anthropic
    #578.34
    Claude Opus 4.8 is #5 with a score of 78.34.
    Compare
  6. Muse Spark 1.1
    Meta
    #677.44
    Muse Spark 1.1 is #6 with a score of 77.44.
    Compare
  7. Grok 4.5
    xAI
    #776.72
    Grok 4.5 is #7 with a score of 76.72.
    Compare

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

  1. Coding100%
    Eligible cohort rank #1 of 122Category score 80.8
  2. Agentic95%
    Eligible cohort rank #7 of 119Category score 65.0
  3. Multilingual0%
    Eligible cohort rank #13 of 13Category score 0.0

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #7 of 119Percentile 95thWeight 22%4 benchmarksVerified65.0
CodingRank #1 of 122Percentile 100thWeight 20%3 benchmarksVerified80.8
ReasoningWeight 17%0 benchmarksNot measuredNot measured
KnowledgeRank Not rankedWeight 12%3 benchmarksVerified93.9
MathRank Not rankedWeight 5%1 benchmarkVerified97.6
MultilingualRank #13 of 13Percentile 0thWeight 7%1 benchmarkVerified0.0
MultimodalRank Not rankedWeight 12%3 benchmarksVerified85.1
Inst. FollowingWeight 5%0 benchmarksNot measuredNot measured

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic4 benchmarks
Terminal-Bench 2.0Provider exact
88%Weighted 38%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A reports Mythos 5 at 88.0% on Terminal-Bench 2.1. BenchLM maps this into the existing Terminal-Bench 2 slot until a separate 2.1 key exists.
OSWorld-VerifiedProvider exact
85%Weighted 34%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A reports Mythos 5 at 85.0% on OSWorld-Verified.
BrowseCompProvider exact
88%Weighted 28%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A and Section 8.14.2 report Mythos 5 at 88.0% on single-agent BrowseComp.
ExploitGymBenchmark exact
17.5%Display only
Source: ExploitGym paperProvenance: ExploitGym reports Claude Mythos Preview with Claude Code produced 157 successful exploits out of 898 tasks. BenchLM stores this as a percentage of the full suite.
Coding3 benchmarks
SWE-bench VerifiedProvider exact

Software Engineering Benchmark Verified

95.5%Weighted 16%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Section 8.2 reports Mythos 5 at 95.5% on SWE-bench Verified, averaged over 5 trials.
SWE-bench ProProvider exact
80.3%Weighted 10%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Section 8.2 reports Mythos 5 at 80.3% on SWE-bench Pro, averaged over 5 trials.
Terminal-Bench 2.0Provider exact
88.0%Display only
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A reports Mythos 5 at 88.0% on Terminal-Bench 2.1. BenchLM stores this in the existing Terminal-Bench display key.
Knowledge3 benchmarks
HLEProvider exact

Humanity's Last Exam

64.5%Weighted 45%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A and Section 8.14.1 report Mythos 5 at 64.5% on Humanity's Last Exam with tools.
GPQAProvider exact

Graduate-Level Google-Proof Q&A

94.1%Weighted 7%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Section 8.8 reports Mythos 5 at 94.1% on GPQA Diamond; BenchLM stores the exact value on the weighted GPQA lane.
HLE w/o toolsProvider exact

Humanity's Last Exam without tools

59%Display only
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A and Section 8.14.1 report Mythos 5 at 59.0% on Humanity's Last Exam without tools.
Math1 benchmark
USAMO 2026Provider exact

United States of America Mathematical Olympiad 2026

97.6%Weighted 10%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Section 8.10 reports Mythos 5 at 99.8% on USAMO 2026 at medium, high, and xhigh reasoning effort, averaged over 10 attempts per problem.
Multilingual1 benchmark
SWE MultilingualProvider exact

SWE-bench Multilingual

92.2%Display only
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Section 8.2 reports Mythos 5 at 92.2% on SWE-bench Multilingual, averaged over 5 trials.
Multimodal3 benchmarks
CharXivProvider exact

CharXiv Reasoning

93.5%Weighted 25%
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A reports Mythos 5 at 93.5% on CharXiv Reasoning with tools.
SWE-bench MultimodalProvider exact
54.9%Display only
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Section 8.2 reports Mythos 5 at 54.9% on SWE-bench Multimodal.
CharXiv w/o toolsProvider exact

CharXiv Reasoning without tools

88.9%Display only
Source: Anthropic: Claude Fable 5 and Claude Mythos 5 system cardProvenance: Table 8.1.A reports Mythos 5 at 88.9% on CharXiv Reasoning without tools.

Frequently Asked Questions

How does Claude Mythos 5 perform overall in AI benchmarks?

Claude Mythos 5 currently ranks #1 out of 200 models on BenchLM's provisional leaderboard with an overall score of 83.93. It also ranks #1 out of 99 on the verified leaderboard. It is created by Anthropic. Its published context window is 1M+.

Is Claude Mythos 5 good for knowledge and understanding?

Claude Mythos 5 has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.

Is Claude Mythos 5 good for coding and programming?

Claude Mythos 5 ranks #1 out of 122 models in coding and programming benchmarks with an average score of 80.8. It is among the top performers in this category.

Is Claude Mythos 5 good for mathematics?

Claude Mythos 5 has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is Claude Mythos 5 good for agentic tool use and computer tasks?

Claude Mythos 5 ranks #7 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 65. It is among the top performers in this category.

Is Claude Mythos 5 good for multimodal and grounded tasks?

Claude Mythos 5 has visible benchmark coverage in multimodal and grounded tasks, but BenchLM does not currently assign it a global category rank there.

Is Claude Mythos 5 good for multilingual tasks?

Claude Mythos 5 ranks #13 out of 13 models in multilingual tasks benchmarks with an average score of 0. There are stronger options in this category.

Does Claude Mythos 5 have full benchmark coverage on BenchLM?

Not yet. Claude Mythos 5 currently has 15 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of Claude Mythos 5?

Claude Mythos 5 has a published context window of 1M+, which determines how much text it can process in a single interaction.

Last updated: July 23, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.