Anthropic · Released Jun 9, 2026
Claude Fable 5
Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...
- #1OTIS Mock AIME
- #1ARC-AGI-1
- #1SimpleBench
- Top 4%SciCode
- Released
- Jun 9 2026
- Context window
- 1M 128K out
- Input / 1M tokens
- $10.00 from $3.00
- Output / 1M tokens
- $50.00 from $18.50
Where to run it
47 providers compared
Included in GitLab Duo. Cheapest: Xpersona at $3.00 / $18.50 per 1M input / output tokens — 74% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| GitLab Duosubscription plan | Plan | Plan | Plan | 1M | 128K | — | Plan |
| Xpersona | $3.00 | $18.50 | $0.30 | 1M | 128K | — | Cheapest |
| Claude Platform on AWS | $10.00 | $50.00 | $1.00 | 1M | 128K | 100%24h | +191% |
| Azure | $10.00 | $50.00 | $1.00 | 1M | 128K | 100%24h | +191% |
| Anthropic | $10.00 | $50.00 | $1.00 | 1M | 128K | 100%24h | +191% |
| Amazon Bedrock | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Googleglobal | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| NanoGPT | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Cloudflare AI Gateway | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Venice AI | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Snowflake Cortex | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Kilo Gateway | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Vercel AI Gateway | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| LLM Gateway | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| FreeModel | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Modelis | $10.00 | $50.00 | — | 1M | 128K | — | +191% |
| 302.AI | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| DigitalOcean | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| ZenMux | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| GitHub Copilot | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Perplexity Agent | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Vertex | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Prime Inference | $10.00 | $50.00 | — | 1M | 128K | — | +191% |
| Neon | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Azure Cognitive Services | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Vivgrid | $10.00 | $50.00 | $1.25 | 1M | 128K | — | +191% |
| Abacus | $10.00 | $50.00 | — | 1M | 128K | — | +191% |
| CrossModel | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Tempr Gateway | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| DevPass (LLM Gateway) | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| CommandCode | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Merge Gateway | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Ofox | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Impossibl | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| OrcaRouter | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| OpenCode | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| NEAR AI Cloud | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Deep Infra | $10.00 | $50.00 | — | 1M | 128K | — | +191% |
| Vertex (Anthropic) | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Eden AI | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Opper | $10.00 | $50.00 | $1.00 | 1M | 128K | — | +191% |
| Standard Compute | $10.00 | $50.00 | $1.00 | 1M | 50K | — | +191% |
| Googleeurope | $11.00 | $55.00 | $1.10 | 1M | 128K | — | +220% |
| Pioneer | $11.00 | $55.00 | $1.10 | 1M | 128K | — | +220% |
| Requesty | $11.00 | $55.00 | $1.10 | 1M | 128K | — | +220% |
| AIHubMix | $11.00 | $55.00 | $1.10 | 1M | 128K | — | +220% |
| Kenari | $12.00 | $57.14 | $1.20 | 1M | 128K | — | +239% |
| AI/ML API | $13.00 | $65.00 | $1.30 | 1M | 128K | — | +278% |
Value
Price vs. performance
Claude Fable 5 Other Anthropic Other labs Value frontier · GPQA Diamond vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
28 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| FrontierSWESoftware engineering · Ultra-long-horizon engineering tasks with a 20-hour budget each | 47.0% | #7of 19 | GPT-6 Astra65.5% |
| DeepSWESoftware engineering · Original long-horizon tasks across active open-source repos | 69.9% | #5of 26 | GPT-6 Astra74.1% |
| FrontierCodeSoftware engineering · Mergeable fixes for hard open-source issues | 53.5% | #2of 41 | Claude Opus 5.554.6% |
| MirrorCodeSoftware engineering · Reimplement whole programs end-to-end without the source | 63.9% | #3of 10 | Claude Opus 5.577.4% |
| GBAEvalSoftware engineering · Build a Game Boy Advance emulator from scratch | 74.5% | #2of 23 | Claude Opus 579.6% |
| GSOSoftware engineering · Optimize real code to make it significantly faster | 78.4% | #3of 27 | Claude Fable 5.188.2% |
| SciCodeSoftware engineering · Research-coding problems from the natural sciences | 61.0% | #4of 102 | Claude Opus 5.566.9% |
| WeirdML v3Software engineering · Unfamiliar ML and data tasks with vague goals | 15.6% | #9of 22 | GPT-6 Astra42.2% |
| APEX-AgentsAgents & computer use · Long professional tasks in banking, consulting and law | 63.6% | #10of 51 | Gemini 4 Argon82.4% |
| Remote Labor IndexAgents & computer use · Real paid freelance projects done end-to-end | 16.1% | #4of 13 | Claude Opus 5.521.3% |
| Vending-Bench 2Agents & computer use · Run a simulated vending business for a year (final balance) | $5,680 | #23of 61 | GPT-6 Astra$15,515 |
| PostTrainBenchAgents & computer use · Post-train small models under a fixed compute budget | 41.8% | #1of 11 | This model |
| GDP.pdfWriting & content · Answer questions about real professional PDFs | 30.0% | #6of 38 | GPT-6.1 Sol35.0% |
| SimpleQA VerifiedWriting & content · Factual accuracy on 1,000 short questions | 70.7% | #7of 79 | Gemini 4 Argon76.0% |
| WebDev ArenaDesign & visual · People vote on which model builds the better website (Elo) | 1628 | #13of 90 | Claude Opus 5.51820 |
| Blueprint-Bench 2Design & visual · Turn apartment photos into 2D floor plans | 38.6% | #6of 32 | Gemini 4 Argon54.4% |
| Furniture AssemblyDesign & visual · Spot mistakes in photos of IKEA builds | 35.8% | #19of 34 | Claude Opus 5.583.3% |
| GPQA DiamondReasoning & knowledge · PhD-level biology, chemistry and physics questions | 85.9% | #57of 122 | Gemini 4 Argon96.4% |
| ARC-AGI-2Reasoning & knowledge · Few-shot abstract reasoning on grid puzzles | 89.2% | #9of 75 | GPT-6 Astra95.0% |
| ARC-AGI-1Reasoning & knowledge · Pattern induction from input–output examples | 98.5% | #1of 75 | Gemini 4 Argon98.5% |
| SimpleBenchReasoning & knowledge · Common-sense trick questions about space, time and people | 81.9% | #1of 58 | This model |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 28.6% | #12of 111 | GPT-5.6 Sol32.3% |
| Chess PuzzlesReasoning & knowledge · Find the single best move in 100 new puzzles | 47.0% | #23of 83 | GPT-6 Astra76.0% |
| LMCAReasoning & knowledge · Judge the quality of conceptual arguments like experts | 61.2% | #8of 123 | Claude Sonnet 5.568.3% |
| FrontierMathMaths · Expert-written problems, undergrad to early research (tiers 1–3) | 87.0% | #9of 87 | GPT-6.1 Sol93.7% |
| FrontierMath Tier 4Maths · Exceptionally hard research-level problems | 90.2% | #4of 64 | GPT-6.1 Sol100.0% |
| ProofBenchMaths · Graduate maths proofs verified in Lean 4 | 95.0% | #8of 70 | Claude Sonnet 5.5100.0% |
| OTIS Mock AIMEMaths · Competition-style problems harder than MATH Level 5 | 100.0% | #1of 120 | Gemini 4 Argon100.0% |
Specs
Details
- Model ID
anthropic/claude-fable-5- Developer
- Anthropic
- Release date
- Jun 9, 2026 ·
- Context window
- 1,000,000 tokens
- Max output
- 128,000 tokens
- API price
- $10.00 / $50.00 per 1M input / output tokens
- Providers
- 47
- Input types
- text, image, file
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- No
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.