AA-OmniscienceVerified
67%
AA-Omniscience Accuracy
Models / Anthropic
Claude Fable 5.1
Claude Fable · Released 2026-09-01
Anthropic's 1 Sep 2026 Mythos-class refresh of Fable 5. Same $10 / $50 per 1M tokens, cache reads $0.25 per 1M. 1M context, 128K output.
Long-horizon coding, multistep research, and dense document work when Opus 5 at higher effort still falls short. Prefer sourced ledger rows over vendor Terminal-Bench-Science tables.
Composite77
Context1M
Input / 1M$10
Output / 1M$50
Knowledge cutoff2026-06-01
Output speed67 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
80/100 avg
- Knowledge
- 67
- Reasoning
- 94
- Coding
- 63
- Agentic
- 91
- Long context
- 85
5 tested benchmark families
Editor's note
Recorded from the 1 Sep 2026 Anthropic launch post and API overview. Editorial depth still needs a desk pass.Benchmark placements
Per-benchmark positions
Where Claude Fable 5.1 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| AA-Omniscience | Knowledge | #1/ 12 | 67 | Artificial Analysis — AA-Omniscience | 2026-09-01 |
| Terminal-Bench | Agentic | #1/ 167 | 91 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #2/ 186 | 59 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #2/ 97 | 63 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #3/ 184 | 85 | Artificial Analysis | 2026-09-28 |
| GPQA Diamond | Reasoning | #8/ 184 | 94 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #91/ 195 | Artificial Analysis | 2026-10-12 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Claude Fable 5.1 shows up, it's highlighted.
KnowledgeClaude Fable 5.1 · 67
- 167
- 265
- 361
ReasoningClaude Fable 5.1 · 94
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingClaude Fable 5.1 · 63
- 174
- 273
- 370
AgenticClaude Fable 5.1 · 91
- 191
- 289
- 389
Long contextClaude Fable 5.1 · 85
- 189
- 286
- 385
Newest receipts
7 recentLast 120 days
Benchmark rows added to the public ledger for Claude Fable 5.1 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 67 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 85.33% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 93.7% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 59.1% | Artificial Analysis | 24 |
| SciCode | Coding | 63.1% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 91.39% | Artificial Analysis | 24 |
| AA-Omniscience | Knowledge | 67% | Artificial Analysis — AA-Omniscience | 51 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Claude Fable 5.1 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Claude Fable 5.1. Scores link to the original source; gaps mean no public row exists yet.
7 of 13 catalog benchmarks have a sourced row for Claude Fable 5.1.
Knowledge
Reasoning
GPQA DiamondVerified
93.7%
GPQA Diamond (AA run)
Humanity's Last ExamVerified
59.1%
HLE (AA run)
Coding
SciCodeVerified
63.1%
SciCode (AA run)
Agentic
Terminal-BenchVerified
91.39%
Terminal-Bench v2.1 (AA run)
Long context
AA-LCRVerified
85.33%
AA-LCR (AA run)
Performance
AA output speedVerified
67 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- Re-checked Anthropic pricing: Fable 5.1 still $10 / $50 per 1M tokens; cache reads $0.25; cache writes $12.50. Consumer Pro still $20/mo or $17 annual; Max from $100.
- Claude Fable 5.1 GA on the Claude API, Bedrock, Vertex, and Microsoft Foundry. Cache reads cut to $0.25 / 1M tokens.
Sources
Where the numbers come from
- Anthropic announcement· checked 2026-09-18
- Claude Fable 5.1 product page· checked 2026-09-18
- Anthropic pricing· checked 2026-09-18
- Claude API Fable 5.1 overview· checked 2026-09-01
- Vals model page· checked 2026-09-05
- Artificial Analysis Claude Fable 5.1· checked 2026-09-10
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Claude Fable 5.1?
Claude Fable 5.1 is a Anthropic model: Anthropic's 1 Sep 2026 Mythos-class refresh of Fable 5. Same $10 / $50 per 1M tokens, cache reads $0.25 per 1M. 1M context, 128K output.
What is Claude Fable 5.1 for?
Long-horizon coding, multistep research, and dense document work when Opus 5 at higher effort still falls short. Prefer sourced ledger rows over vendor Terminal-Bench-Science tables.
What does Claude Fable 5.1 cost?
Claude Fable 5.1 lists $10 input and $50 output per 1M tokens. Pricing last checked 2026-09-18.
How large is the Claude Fable 5.1 context window?
Claude Fable 5.1 publishes a 1M token context window.
What is the VerdictPal status of Claude Fable 5.1?
Claude Fable 5.1 carries Solid status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.