Claude Fable 5
Claude Fable 5 is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.
Use it for
- Teams evaluating coding, rag, and agents
- Workloads that can use a 1m context window
- Buyers comparing 4 tracked provider routes
Do not use it for
- Workloads where another current model has stronger sourced task evidence
- Family
- Claude Fable
- Released
- 2026-06-09
- Context
- 1m
- Max output
- 128,000
- Architecture
- Decoder Only
- Specialization
- general
- Openness
- Proprietary
- License
- ProprietaryCommercial use: conditional
- Weights
- Not released
- Code
- Unknown
- Training
- Pretrained
Cheapest of 8 routes · Anthropic · cache read $1.00
About
Anthropic's most capable widely released model, built for demanding reasoning and long-horizon agentic work. Claude Fable 5 is the generally available Mythos-class Claude model, supports vision, tool use, structured outputs, prompt caching, Batch API processing, adaptive thinking that is always on, a 1M-token context window, and up to 128k output tokens. It launched on the Claude API, AWS Bedrock, Vertex AI, and Microsoft Foundry on June 9, 2026 with first-party pricing at $10 per 1M input tokens and $50 per 1M output tokens. Anthropic temporarily disabled Fable 5 access on June 12, 2026 after a US export control directive and redeployed Fable 5 and Mythos 5 on July 1, 2026.
Claude Fable 5 is a proprietary model in the Claude Fable family. The structured metadata tracks a 1m-token context window, multimodal input, reasoning, function calling, tool use, and structured outputs. This page tracks provider routes through Anthropic, AWS Bedrock, GCP Vertex AI, and 5 more, with the cheapest tracked route listed at $10 input and $50 output per 1M tokens. Headline tracked benchmarks include SWE-bench Pro 80.3, SWE-bench Verified 96.0, and OSWorld-Verified 85.0.
Top use-case fit: coding, agents, and build tasks
Coding
Q/$ D2 relevant benchmarks in the decision map.
RAG
Included by capability and metadata signals in the decision map.
Agents
Q/$ D1 relevant benchmark in the decision map.
Provider price ladder
Compare all 8Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Batch in / out | Cache | Route |
|---|---|---|---|---|---|
| Anthropic | $10.00 | $50.00 | $5.00 / $25.00 | read $1.00 / 5m $12.50 / 1h $20.00 | Serverless |
| AWS Bedrock | $10.00 | $50.00 | - | - | Serverless |
| GCP Vertex AI | $10.00 | $50.00 | - | read $1.00 / 5m $12.50 | Serverless |
| Microsoft Foundry | $10.00 | $50.00 | - | - | Serverless |
Available via routers & gateways(16)
LiteLLM
GatewayOpen-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.
OpenRouter
HybridUnified hybrid gateway to 400+ models from 60+ providers via a single OpenAI-compatible API, with optional auto-routing that selects the best model per prompt.
Portkey
GatewayProduction AI gateway routing to 1,600+ LLMs with failover, load balancing, semantic caching, and guardrails; Apache 2.0 core is fully self-hostable with the complete feature set.
AIRouter
RouterCommercial LLM router that analyzes incoming requests and routes to the optimal model for cost/quality/latency via a drop-in OpenAI-compatible API, with a privacy-preserving embedding mode that avoids sending prompt content.
Amazon Bedrock Intelligent Prompt Routing
RouterAWS Bedrock's native intelligent prompt router that routes prompts between Anthropic Claude model tiers (Haiku/Sonnet) based on predicted task complexity, with no extra per-routing charge.
Azure AI Foundry Model Router
RouterMicrosoft Azure AI Foundry's native model router that uses a trained ML model to route each prompt in real time to the optimal Azure-hosted model, with Balanced/Cost/Quality mode selection and automatic failover.
Capabilities
Benchmark peer barsfor Coding
Benchmark scores(14)
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| SWE-bench Pro | 80.3 | SWE-bench ProObserved 2026-06-09 | — | Source |
| SWE-bench Verified | 96.0 | Vals.ai harness (weighted avg by difficulty tier)Observed 2026-06-09 | — | Source |
| OSWorld-Verified | 85.0 | OSWorld-VerifiedObserved 2026-06-09 | — | Source |
| GDPval-AA | 1932.0 | GDPval-AA ELOObserved 2026-06-09 | — | Source |
| Blueprint-Bench 2 | 38.6 | Blueprint-Bench 2Observed 2026-06-09 | — | Source |
| AutomationBench | 17.4 | AutomationBenchObserved 2026-06-09 | — | Source |
| Legal Agent Benchmark | 13.3 | Legal Agent BenchmarkObserved 2026-06-09 | — | Source |
| GDP.pdf | 29.8 | GDP.pdf vision, no toolsObserved 2026-06-09 | — | Source |
| CursorBench | 72.9 | CursorBench 3.1Observed 2026-06-30 | Configuration: Fable 5 Max Harness: CursorBench 3.1 Evaluator: Cursor Confidence: confirmed Notes: Highest CursorBench 3.1 score across Cursor's published effort configurations for this base model. | Source |
| CursorBench | 70.5 | CursorBench 3.2Observed 2026-07-18 | Configuration: Fable 5 Max Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $17.32 Tokens/task: 103,525 Steps/task: 72 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful. | Source |
| CursorBench | 68.4 | CursorBench 3.2Observed 2026-07-18 | Configuration: Fable 5 Extra High Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $11.73 Tokens/task: 64,971 Steps/task: 56 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful. | Source |
| CursorBench | 66.5 | CursorBench 3.2Observed 2026-07-18 | Configuration: Fable 5 High Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $8.77 Tokens/task: 43,747 Steps/task: 48 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful. | Source |
| CursorBench | 65.2 | CursorBench 3.2Observed 2026-07-18 | Configuration: Fable 5 Medium Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $6.80 Tokens/task: 30,366 Steps/task: 41 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful. | Source |
| CursorBench | 62.1 | CursorBench 3.2Observed 2026-07-18 | Configuration: Fable 5 Low Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $4.46 Tokens/task: 18,182 Steps/task: 31 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful. | Source |
Migration checks
No linked migration route is available for this model yet.
Rankings & picks(3)
Comparison and alternatives
Browse all comparisons →Frequently asked questions
What is the context window of Claude Fable 5?
Claude Fable 5 has a context window of 1m tokens.
What is the max output of Claude Fable 5?
Claude Fable 5 can generate up to 128,000 output tokens.
How much does Claude Fable 5 cost?
Claude Fable 5 is available at $10.00/1M input tokens through Anthropic.
When was Claude Fable 5 released?
Claude Fable 5 was released on 2026-06-09.
Which providers offer Claude Fable 5?
Claude Fable 5 is available from 8 providers: Anthropic, AWS Bedrock, GCP Vertex AI, Microsoft Foundry, OpenRouter, Vercel AI Gateway, Replicate API, Databricks Foundation Model Serving.
What benchmarks has Claude Fable 5 been tested on?
Claude Fable 5 has been evaluated on 14 benchmarks, including SWE-bench Pro, SWE-bench Verified, OSWorld-Verified, GDPval-AA, Blueprint-Bench 2.
Cheapest of 8 routes · Anthropic · cache read $1.00