Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Anthropic · Proprietary · bench-align-v5
Claude Mythos 5.2 by Anthropic costs $8.00 per 1M input tokens and $40.00 per 1M output tokens and carries a context window of 1M tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Every published price component, including cache reads and writes.
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
Lifecycle events the source associates with this model, plus what this profile does not claim.
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: ObservedUnavailable
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | Unavailable | Unavailable | Unavailable | |
| Throughput | Unavailable | Unavailable | Unavailable |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
Plans from the model's maker, and the API providers that list this exact model today. Links leave ModelScale; each provider sets its own prices and limits.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| SingleCellBench LatchBio SingleCellBench | 65.6 | #1 | 195 single-cell RNA sequencing problems | Professional bioinformatics | Claude Opus 5 System Card |
| Protein Design Anthropic Protein Design evaluation | 64.6 | #1 | Constrained protein-sequence design tasks | Computational protein design | Claude Opus 5 System Card |
| Protocols (troubleshooting) Molecular Biology Protocols Troubleshooting | 77.1 | #1 | Molecular-biology protocol troubleshooting | Expert laboratory protocols | Claude Opus 5 System Card |
| Morphology-to-molecule matching Axiom Bio morphology-to-molecule matching | 35.2 | #1 | Four multi-channel cell-imaging perturbation-matching tasks | Biomedical image and molecular reasoning | Claude Opus 5.5 System Card |
| Medicinal chemistry Anthropic medicinal-chemistry evaluation | 69.0 | #1 | 504 molecular-property questions | Medicinal chemistry reasoning | Claude Opus 5.5 System Card |
| De novo protein-binder design Anthropic de novo protein-binder design evaluation | 85.1 | #1 | 15 protein targets | Long-horizon computational protein design | Claude Opus 5.5 System Card |
| Protein Mutation Attribution Protein mutation attribution | 51.5 | #1 | Protein mutation questions | Protein understanding | Claude Fable 5.2 and Claude Mythos 5.2 system card |
| Protein Domain ID Protein domain identification | 91.8 | #1 | Multiple-choice Pfam family questions | Protein understanding | Claude Fable 5.2 and Claude Mythos 5.2 system card |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Terminal-Bench 4.0 Terminal-Bench 4.0 | 70.50 | #2 | 66 professional computer-work tasks | Frontier autonomous knowledge work | Terminal-Bench 4.0 |
| Terminal-Bench-Science 0.1 Terminal-Bench-Science 0.1 | 75.3 | #2 | 70 expert-curated scientific research workflows | Frontier scientific research workflows | Terminal-Bench-Science 0.1 |
| ExploitGym ExploitGym | 38.1 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| ExploitBench ExploitBench v8-bench | 95 | #7 | V8 exploit synthesis runs | Browser exploitation and cybersecurity | ExploitBench |
| CyScenarioBench success CyScenarioBench Average Success Rate | 76 | #1 | 11 cyber scenarios | Long-horizon cybersecurity | Assessing GPT-5.6 Sol |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Biomedical image analysis Anthropic biomedical-image-analysis evaluation | 80.7 | #1 | 91 cases across 31 biomedical imaging tasks | Biomedical image analysis | Claude Opus 5.5 System Card |
| #5 |
| 898 exploitation tasks |
| Advanced cybersecurity exploitation |
| ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? |
| Binary Exploitation Benchmark Anthropic Binary Exploitation Benchmark (formerly OSS-Fuzz), control-flow hijacks | 148 | #1 | Open-source function entry points | Exploit development | Claude Fable 5.2 and Claude Mythos 5.2 system card |
| CVE Rediscovery Anthropic CVE Rediscovery | 89.6 | #1 | 1,829 CVE advisories | Vulnerability discovery | Claude Fable 5.2 and Claude Mythos 5.2 system card |