Skip to content

Claude Opus 4.7 (Adaptive)

AnthropicProprietaryReasoningProvisionalReleased Apr 16, 2026
Compare with GPT-5.4 ProSource scores ↗
Overall score
68.7
Rank 23 of 222 · 54.3–83.0
Context window
1M
1,000,000 tokens
Price per 1M tokens
$5.00 in · $25.00 out
Source: BenchLM
Blended price
$10.00
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Knowledge#3082.0
  2. 2Agentic & tool use#2971.8
  3. 3Coding#3265.6
  4. 4Reasoning57.4
  5. 5Multimodal#4753.7
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (19)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Agentic & tool use

BrowseComp79.3#23 of 50
MCP Atlas77.3#13 of 42
Terminal-Bench 2.069.4#6 of 39
OSWorld-Verified78.0#14 of 35
JobBench45.9#15 of 34
CyberGym73.1#18 of 25
OSWorld 2.018.2#17 of 24

Reasoning

ARC-AGI-275.8#18 of 41
ARC-AGI-30.2#18 of 22
MRCR v2 128K-256K59.2#2 of 2

Multimodal

CharXiv91.0#5 of 42
CharXiv w/o tools82.1#10 of 14
OfficeQA Pro43.6#13 of 13

Knowledge

GPQA94.2#3 of 117
HLE54.7#12 of 69
HLE w/o tools46.9#9 of 43

Coding

SWE-bench Verified87.6#6 of 98
SWE-bench Pro64.3#15 of 69

Math

FrontierMath (legacy)43.8#7 of 7

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (BenchLM)$5.00$25.00—

Anthropic official pricing for Claude Opus 4.7 applies to the adaptive reasoning variant as the same API model with effort controls; launch announcement states pricing remains $5 input / $25 output per million tokens.

Speed

Measured by us through OpenRouter (median of trials).

Not available on OpenRouter, so we cannot measure it yet.

Related models

Same family

Similar score