MDL
Claude Fable 5
Anthropic
Verified against Artificial Analysis · Jul 27, 2026
Capability
95
percentile index
Price /M
$10 / $50
in / out
Context
1M
tokens
Avg score
71.6
7 benchmarks
AA Intelligence
59.9
index
Speed
58 tok/s
TTFT 52s
Released
Jun 9, 2026
About
Claude Fable 5 is a proprietary model in the Claude family from Anthropic. Benchmarked on 7 evals, averaging 71.6.
Benchmark scores · 7
avg 71.6 — higher is betterThe research behind this model
all papers mentioning it →An Exam for Active Observers▲ 25 on HF · Jul 17, 2026LoSoNA: A Benchmark for Local Social Norm Adaptation in Group Conversations▲ 6 on HF · Jun 12, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed58 tok/s
Latency (TTFT)52s
Price in / out$10 / $50 per 1M
Blended price$20 / 1M
Coding index76.5
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
GPQA Diamondfirst seen at 92.6
1mo ago
Humanity's Last Examfirst seen at 53.3
1mo ago
SciCodefirst seen at 60.2
1mo ago
IFBenchfirst seen at 63.5
1mo ago
τ²-benchfirst seen at 98.5
1mo ago
Terminal-Bench Hardfirst seen at 62.9
1mo ago
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$50/M