Anthropicproprietary
Released: 2026-06-09

Claude Fable 5

Anthropic flagship Mythos-class model delivering top-ranked reasoning, autonomous agentic work, and codebase-wide software migration.

Intelligence Index
96.0/ 100
Calibrated multi-domain composite
Throughput Speed
45tok/sec
Output streaming throughput
Time To First Token
480ms
Initial chunk server latency
LiveBench Contam-Free
83.0%
September 2026 suite score

LiveBench Multi-Domain Evaluation Breakdown

Monthly refreshed contamination-free test sets
Logical Reasoning89.7%
Coding & Repo Repair86.0%
Mathematics (AIME)96.0%
Data Analysis & Tables81.2%
Language & Comprehension75.4%
Instruction Following70.0%

Sourced Benchmark Evaluations (3)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
ARC-AGI-2Nearing Saturation75.3%
75
2026-07-07independentSource ↗
SWE-bench VerifiedSaturated65.4%
25
2026-06-18independentSource ↗
IFEvalDeprecated70%
25
2026-06-13independentSource ↗
Anthropic Python SDKAPI Ready
import anthropic

client = anthropic.Anthropic()
message = client.messages.create(
    model="claude-fable-5",
    max_tokens=2048,
    temperature=0,
    messages=[{"role": "user", "content": "Execute SWE-bench verified repository fix"}]
)
print(message.content[0].text)

Evaluation & Sourcing Notes

Scores listed for Claude Fable 5 represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.

All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.