Data as of October 20, 2026 · How the score is built
Claude Fable 5
Decision readingClaude Fable 5 scores 81.4 out of 100 and ranks #7 of 227. This profile shows 25 source-displayable benchmark rows; its strongest eligible category is Agentic at #6. API pricing is $10 input and $50 output per million tokens, with cached input at $1.
Released Jun 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
81.4/100
field median 49.8#7 of 227 ranked models
Public
#7of 227
Verified #7 of 87
Price
$10input / $50 output
input median $0.72cached $1 · batch + cache $0.50 · blended $30
Speed
68tok/s
field median 93.5 tok/sFirst token 100.13 s
Context
1Mtokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Agentic ranks #6. Particularly useful for coding agents, browser research, and computer-use workflows.
Validate before choosing
25 published rows leave some tracked benchmark slots empty. Coding is its lowest eligible category at #7.
Source-linked · 25 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 |
|---|---|---|---|---|---|---|
| AgenticRank #6 of 131Percentile 96thWeight 22%6 benchmarksVerified | 75.8 | 6 benchmarks | Verified | |||
| CodingRank #7 of 152Percentile 96thWeight 20%10 benchmarksVerified | 71.3 | 10 benchmarks | Verified | |||
| ReasoningRank Not rankedWeight 17%2 benchmarksVerified | 84.3 | 2 benchmarks | Verified | |||
| MultimodalRank Not rankedWeight 12%3 benchmarksVerified | 66.0 | 3 benchmarks | Verified | |||
| KnowledgeRank #6 of 187Percentile 97thWeight 12%2 benchmarksVerified | 82.6 | 2 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 | |||
| MathRank Not rankedWeight 5%2 benchmarksVerified | 96.0 | 2 benchmarks | Verified |
25 of 726 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.
- Agentic6/6 verified
- Coding10/10 verified
- Reasoning2/2 verified
- Multimodal3/3 verified
- Knowledge2/2 verified
- MultilingualNot measured
- Inst. FollowingNot measured
- Math2/2 verified
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Claude Fable 5 category percentile values
- Agentic96th percentile
- Coding96th percentile
- ReasoningNot eligible
- MultimodalNot eligible
- Knowledge97th percentile
- MultilingualNot eligible
- Instruction followingNot eligible
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#6/131
- Coding#7/152
- ReasoningNot ranked
- MultimodalNot ranked
- Knowledge#6/187
- MultilingualNot ranked
- Inst. FollowingNot ranked
- MathNot ranked
Benchmark ledger
Coding 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.
Coding10 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench Pro | Score80% | Versus best verified row Best verified: Claude Opus 5.5 · 89.9% | Gap9.9 behind | Weight26% ref. weight | |
| CursorBench 3.2 | Score70.5% | Versus best verified row Best verified: Claude Fable 5.1 · 73.4% | Gap2.9 behind | Weight10% ref. weight | Benchmark exact |
| FrontierSWE v2 | Score47.0% | Versus best verified row Best verified: GPT-6 Astra · 65.5% | Gap18.5 behind | Weight8% ref. weight | Benchmark exact |
| LiveCodeBench (Vals)LiveCodeBench, Vals AI run | Score89.8% | Versus best verified row Best verified: Claude Fable 5.1 · 90.5% | Gap0.7 behind | Weight8% ref. weight | |
| FrontierCode 1.1 Main | Score53.5% | Versus best verified row Best verified: Claude Opus 5.5 · 54.4% | Gap0.9 behind | Weight4% ref. weight | Benchmark exact |
| VulcanBench v3 | Score89.5% | Versus best verified row Best verified: Grok 4.5 · 89.9% | Gap0.4 behind | Weight3% ref. weight | Benchmark exact |
| Terminal-Bench 2.1 | Score84.3% | Versus best verified row Best verified: SWE-2 · 92.8% | Gap8.5 behind | WeightScored in Agentic | |
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score95% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap1 behind | WeightDisplay only | |
| cursorBench31 | Score70.6% | Versus best verified row Best verified: Claude Fable 5 · 70.6% | GapBest verified | WeightDisplay only | Benchmark exact |
| SWE-bench (Vals)SWE-bench, Vals AI run | Score95.0% | Versus best verified row Best verified: Claude Opus 5 · 97.0% | Gap2 behind | WeightDisplay only | Verified |
Agentic6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 2.1 | Score84.3% | Versus best verified row Best verified: SWE-2 · 92.8% | Gap8.5 behind | Weight8% ref. weight | |
| OSWorld-Verified | Score85% | Versus best verified row Best verified: Qwen3.8 Max · 86.1% | Gap1.1 behind | Weight7% ref. weight | |
| Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run | Score80.5% | Versus best verified row Best verified: GPT-6 Astra · 87.3% | Gap6.8 behind | Weight4% ref. weight | |
| Terminal-Bench 3.0 | Score34.0% | Versus best verified row Best verified: Claude Opus 5 · 42.7% | Gap8.7 behind | Weight3% ref. weight | Benchmark exact |
| ApprenticeBenchApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job | Score34% | Versus best verified row Best verified: Claude Fable 5.1 · 72% | Gap38 behind | WeightDisplay only | Benchmark exact |
| MazeBenchMazeBench: visual spatial reasoning in a 3D open world | Score11% | Versus best verified row Best verified: GPT-6 Astra · 23% | Gap12 behind | WeightDisplay only | Benchmark exact |
Reasoning2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| ARC-AGI-2Abstraction and Reasoning Corpus for AGI v2 | Score89.2% | Versus best verified row Best verified: GPT-6 Astra · 95% | Gap5.8 behind | WeightWeighted 25% | Benchmark exact |
| ARC-AGI-1ARC-AGI-1 Semi-Private Evaluation | Score98.50% | Versus best verified row Best verified: Claude Fable 5 · 98.50% | GapBest verified | WeightDisplay only | Benchmark exact |
Multimodal3 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| OfficeQA Pro | Score57.9% | Versus best verified row Best verified: Claude Opus 5.5 · 67.7% | Gap9.8 behind | WeightWeighted 25% | |
| ParseBenchParseBench: document parsing for AI agents (LlamaIndex leaderboard) | Score70.78% | Versus best verified row Best verified: Claude Opus 5.5 · 79.85% | Gap9.1 behind | WeightDisplay only | |
| Blueprint-Bench 2 | Score38.6% | Versus best verified row Best verified: Claude Fable 5 · 38.6% | GapBest verified | WeightDisplay only |
Knowledge2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMLU-Pro (Vals)MMLU-Pro, Vals AI run | Score91.5% | Versus best verified row Best verified: Claude Fable 5.1 · 92.4% | Gap0.9 behind | Weight6% ref. weight | Verified |
| GPQA Diamond (Vals)GPQA Diamond, Vals AI run | Score93.2% | Versus best verified row Best verified: Gemini 3.1 Pro · 95.5% | Gap2.3 behind | Weight2% ref. weight |
Math2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)FrontierMath v2 Tiers 1-3 | Score87.018% | Versus best verified row Best verified: GPT-6 Astra · 93.684% | Gap6.7 behind | WeightWeighted 30% | Benchmark exact |
| FrontierMath v2 (Tier 4)FrontierMath v2 Tier 4 | Score90.200% | Versus best verified row Best verified: GPT-6.1 Sol · 100.000% | Gap9.8 behind | WeightWeighted 10% | Benchmark exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 25 rowsWhat it costs to get this score
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
Claude Fable 5 · 81.3 score · $30 blended per million tokens
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
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.
Base entry
Claude Fable 5.1 · 83.3Spec 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-fable-5Anthropic model overview
- Context window
- 1MAnthropic model overview
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, imageAnthropic model overview
- Output modalities
- textAnthropic model overview
- Parameters
- Not disclosed by the provider
- Availability
- Claude APIAnthropic model overview
- Cloud regions
- Not tracked yet
- Lifecycle
- activeAnthropic model overview
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $1 per million cached input tokensClaude API pricing
- 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.
Claude Fable 5 ranks #7 of 227 on the public leaderboard with a score of 81.35/100. Its source-verified position is #7 of 87.
Claude Fable 5 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.
Claude Fable 5 is Anthropic's generally available Mythos-class model. It uses the same underlying model as Claude Mythos 5, with production safeguards for cybersecurity, biology, chemistry, distillation, and frontier-AI-development requests. Some Claude clients route flagged requests to Claude Opus 4.8; the Messages API blocks by default unless fallback is configured.
Updated from Anthropic's June 9, 2026 Claude Fable 5 and Claude Mythos 5 launch pages and system card, supplemented by CursorBench v3.1, CursorBench v3.2's Fable 5 Max setting, and the Fable 5.1 system card's separate FrontierSWE v2 comparison. Anthropic says Fable 5 and Mythos 5 use the same underlying model, with Fable adding production safeguards and fallback/block behavior. We keep the 34-task FrontierSWE v2 mean separate from the original 17-task FrontierSWE dominance lane.
Claude Fable 5 sits in the Claude Fable family with Claude Fable 5.1. 25 of 726 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Agentic at #6, while its lowest eligible position is Coding at #7. particularly useful for coding agents, browser research, and computer-use workflows.
Last updated October 20, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Where to run Claude Fable 5 by API
2 hosts publish a per-token price for it. DeepInfra has the lowest output rate, $50.00 per million tokens. Prices read from each host’s own pages on October 20, 2026.
| Host | Input /1M | Output /1M | Cached /1M | Context |
|---|---|---|---|---|
| DeepInfra | $10.00 | $50.00 | — | 1M |
| OpenRouter | $10.00 | $50.00 | $1.00 | 1M |
Questions
How does Claude Fable 5 perform overall in AI benchmarks?
Claude Fable 5 ranks #7 out of 227 models on the public BenchAlign leaderboard, with a score of 81.35/100. Its evidence status is Supported, and this profile shows 25 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Claude Fable 5 good for knowledge and understanding?
Claude Fable 5 ranks #6 out of 187 eligible models for knowledge and understanding, with a public category score of 82.6/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Claude Fable 5 good for coding and programming?
Claude Fable 5 ranks #7 out of 152 eligible models for coding and programming, with a public category score of 71.3/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Claude Fable 5 good for mathematics?
Claude Fable 5 has source-displayable benchmark coverage for mathematics, 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 Fable 5 good for reasoning and logic?
Claude Fable 5 has source-displayable benchmark coverage for reasoning and logic, 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 Fable 5 good for agentic tool use and computer tasks?
Claude Fable 5 ranks #6 out of 131 eligible models for agentic tool use and computer tasks, with a public category score of 75.8/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Claude Fable 5 good for multimodal and grounded tasks?
Claude Fable 5 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 Fable 5?
Claude Fable 5 belongs to the Claude Fable family. Related tracked variants include Claude Fable 5.1. 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 Fable 5 have full benchmark coverage on BenchLM?
No. Claude Fable 5 currently has 46 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 Fable 5?
Claude Fable 5 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.