MDL
Claude Opus 4.6
Anthropic
Verified against Artificial Analysis · Jul 13, 2026
Capability
90
percentile index
Price /M
$5 / $25
in / out
Context
1M
tokens
Avg score
62.9
7 benchmarks
AA Intelligence
43.7
index
Speed
48 tok/s
TTFT 18s
Released
Feb 5, 2026
About
Claude Opus 4.6 is a proprietary model in the Claude family from Anthropic. Benchmarked on 7 evals, averaging 62.9.
Benchmark scores · 7
avg 62.9 — higher is betterThe research behind this model
all papers mentioning it →ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research▲ 87 on HF · May 28, 2026TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents▲ 47 on HF · Jun 26, 2026Streaming Communication in Multi-Agent Reasoning▲ 29 on HF · Jun 3, 2026AutoLab: Can Frontier Models Solve Long-Horizon Auto Research and Engineering Tasks?▲ 28 on HF · Jun 3, 2026OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks▲ 22 on HF · Jun 28, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed48 tok/s
Latency (TTFT)18s
Price in / out$5 / $25 per 1M
Blended price$10 / 1M
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
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$25/M