Inference
EN
Anthropic

Claude Mythos 5

Claude Mythos 5 is a reasoning model from Anthropic in the Claude Mythos family. 10 benchmarks count toward its index, in 4 categories.

  • Closed weights
  • Reasoning
  • Input: text
  • Released 2026-06
RankedEnough results for an index.
IndexAll categories, out of 100.70.8±8.6Median 51.0
CoverageIndex weight that has results.65%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.$10.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$50.00Median $2.55
ContextMaximum tokens in one request.1MMedian 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.4 counted70.025 / 218
CodingCode writing and repair.2 counted74.412 / 209
ReasoningLogic problems and puzzles.None counted——
MultimodalTasks with images and text.2 counted70.810 / 84
KnowledgeFacts and expert knowledge.2 counted74.99 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.None counted——

Results

10 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Software Engineering Benchmark VerifiedCoding95.5%2 / 114
CharXiv ReasoningMultimodal93.5%1 / 49
Graduate-Level Google-Proof Q&AKnowledge94.1%7 / 141
OSWorld-VerifiedAgentic85.0%2 / 37
SWE-bench ProCoding80.3%5 / 83
Humanity's Last ExamKnowledgeAbove fitFar above this model's index.64.5%5 / 77
Humanity's Last Exam without toolsKnowledge59.0%4 / 53
CharXiv Reasoning without toolsMultimodal88.9%2 / 17
Terminal-Bench 2.1 · BenchLMAgentic88.0%11 / 86
BrowseCompAgentic88.0%13 / 63
CyberGymAgentic83.8%11 / 29
United States of America Mathematical Olympiad 2026Math97.6%—

Sources

BenchLMCC BY-NC 4.0

History

5 releases
Index71.8 → 70.8
23 Oct 202612 Nov 2026
Output / 1M$50.00 → $50.00
23 Oct 202612 Nov 2026

Questions

How much does Claude Mythos 5 cost?
$10.00 per 1M input tokens, $50.00 per 1M output tokens.
How good is Claude Mythos 5?
Its index is 70.8 out of 100. That is place 20 of 283 ranked models.
What is Claude Mythos 5 best at?
Knowledge: 74.9 out of 100, place 9 of 268.
What is the context window of Claude Mythos 5?
1M tokens.
Is Claude Mythos 5 open source?
No. Its weights are not published.
When was Claude Mythos 5 released?
June 2026.

Badge

Claude Mythos 5 on Inference360: place 20 of 283, index 70.8
HTML
<a href="https://inference360.ai/models/claude-mythos-5"><img src="https://inference360.ai/badges/claude-mythos-5.svg" alt="Claude Mythos 5 on Inference360: place 20 of 283, index 70.8" height="20"></a>
Markdown
[![Claude Mythos 5 on Inference360: place 20 of 283, index 70.8](https://inference360.ai/badges/claude-mythos-5.svg)](https://inference360.ai/models/claude-mythos-5)

Compare with

MiMo-V2.6-Pro70.2MiniMax M3.1 Flash69.3GPT-5.569.1GPT-5.6 Terra68.9GPT-5.6 Sol73.0

Lower price

DeepSeek V4.1 Flash 101768.5 · $0.40/1MMiniMax M3.1 Flash69.3 · $0.60/1MMiMo-V2.6-Pro70.2 · $0.87/1M

More from Anthropic

All models
Claude Fable 574.0Claude Fable LatestUnrankedClaude Opus 4.867.7Claude Sonnet 564.2Claude Haiku LatestUnranked