none reviewed yet · 2 benchmark observations · latest capture Sep 8, 2026
Inspect the evidenceAnthropic · MODEL PROFILE
Claude Fable 5.1
Senior teams shipping production frontends who want the strongest all-round model and will pay for it
Saved editorial fit for your current task lens; not a benchmark result
How scoring worksVerified rates (Sep 8, 2026) · 100 runs / month at your token settings
Adjust workloadAt a glance
9 facts- Provider
- Anthropic
- Type
- API model
- Access
- Anthropic API, Claude apps, AWS Bedrock; Claude Code harness
- Context
- 1,000,000 tokens
- Maximum output
- 128K tokens
- License / weights
- Proprietary
- Input modalities
- text, image, file
- Output modalities
- text
- Pricing notes
- $10 in / $50 out per 1M; cache read $0.25 (75% cheaper than Fable 5), cache write $12.50; Anthropic reports ~25% lower typical workload cost and up to ~45% savings on complex agentic coding vs Fable 5
The practical trade-offs
Documented strengths
- SWE-bench Pro rank 1 (81.2%) and AA Intelligence Index rank 1 (57)
- Arena WebDev Frontend rank 2 (1796) plus Design Arena Website rank 3 (1330) - strong on both human-preference boards
- Best-documented debugging evidence (Red Hat diagnosed every broken build; Millennium's one-in-a-million crash)
- Cache reads at $0.25 cut agentic loop costs sharply
Limitations to check
- $10/$50 top-tier pricing
- AA measured TTFT of 266.51 s at max effort
- Anthropic's own table shows Fable 5 ahead on Terminal-Bench 4.0 in the Mythos row - effort settings matter
- Arena row has only 1,893 votes
Follow the source
7 source linksEvidence & conditions
7Source links
NoneReviewed records
2Benchmark runs
Sep 8, 2026Latest capture
- Evidence source 1
anthropic.comDate not recorded - Evidence source 2
docs.aws.amazon.comDate not recorded - Evidence source 3
anthropic.comDate not recorded - Evidence source 4
arena.aiDate not recorded - Evidence source 5
benchlm.aiDate not recorded - Evidence source 6
benchlm.aiDate not recorded - Evidence source 7
artificialanalysis.aiDate not recorded
Benchmark observations
Terminal-Bench 4.0 Inspect conditions
- name
- Terminal-Bench 4.0
- value
- 55.8
- unit
- %
- source Url
- https://openai.com/index/gpt-6-astra/
- checked At
- 2026-09-08T19:30:56Z
- attribution
- provider-reported
- configuration
- OpenAI comparative table, not an independent Anthropic verification; maximum across effort; competitor harness not specified.
DeepSWE v1.1 Inspect conditions
- name
- DeepSWE v1.1
- value
- 67.4
- unit
- %
- source Url
- https://openai.com/index/gpt-6-astra/
- checked At
- 2026-09-08T19:30:56Z
- attribution
- provider-reported
- configuration
- OpenAI comparative table, not an independent Anthropic verification; maximum across effort; competitor harness not specified.
Source reports, measured results, and editorial assessments are separate evidence types. Source dates do not imply a new benchmark run. Capture dates are shown when recorded; otherwise we say the capture date was not recorded. Recorded dates are compared with this catalogue snapshot (Sep 11, 2026), not today’s date.
Method AIS-V2
How this page is assessed
Quality claims follow Method AIS-V2.
Money never buys rank — commercial relationships cannot change rank, eligibility, evidence quality, or publication timing.
Submit or claim a listing is free and does not skip the editorial queue.
Private · this device only