C
Claude Opus 5.5
The most capable closed model you can actually use — first on both long-horizon coding and professional reasoning
Intelligence v5
58.4
Leaderboard rank
#2
Input / output
$4.00/$20.00
Blended 3:1
$8.00
Output speed
96 tok/s
Value
7 pts/$
Capability profile Normalised to the field maximum
Specifications
Vendor
Anthropic
Released
2026-10
Weights
Closed, API only
Context
1,000,000 tokens
Max output
128,000 tokens
Parameters
Undisclosed
Input price
$4.00 / 1M
Output price
$20.00 / 1M
Cache hit
$0.20 / 1M
Output speed
96 tok/s
Intelligence
58.4（v5）
Benchmark scores
How it is measured →Ranked 2 · bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 Intelligence index | 58.4 | 98% | −1.0 pts |
| AA Intelligence Index v4.3.2 Intelligence index (legacy) | 57.6 | 100% | |
| GPQA Diamond Graduate-level science reasoning | 93.9 | 97% | −2.5 pts |
| Terminal-Bench 4.0 Terminal-agent tasks | 65.2 | 100% | |
| LiveBench Design Document generation (DOCX/PPT/PDF) | 86.9 | 100% | |
| Vals Index Vals composite index | 72.4 | 99% | −0.7 pts |
Assessment
Ranks first on AA Intelligence Index v5 among all publicly callable models. Joint first on Terminal-Bench 4.0 at 65.2%, and its AA-Omniscience hallucination penalty of 46 leads Sonnet 5.5 by fourteen points. 20% cheaper than Opus 5 ($5/$25 → $4/$20).
At the same price
The four models closest on blended price, side by side.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| Claude Sonnet 5.5Anthropic | 54.2 | $4.00 | 135 | -4.2 |
| GPT-6.1 SolOpenAI | 54.5 | $4.00 | 77 | -3.9 |
| Gemini 4 CarbonGoogle | 59.4 | $4.00 | 87 | +1.0 |
| Gemini 4 ArgonGoogle | 50.4 | $4.00 | 80 | -8.0 |
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard