Noometry

Anthropic, proprietary

Claude Opus 4.7

Claude Opus 4.7 by Anthropic ranks 19th of 354 ranked models on the Noometry Index as of October 2026, with a score of 58.3. Its strongest category is writing & preference, where it ranks 8th. API pricing starts at $5 per million input tokens and $25 per million output tokens, with a 1M-token context window.

Last verified

Noometry rank
#19 of 354
Index score
58.3
Evidence
Confirmed 66 results
Provider
Anthropic
Released
April 14, 2026
Weights
Proprietary
Reasoning
Yes
Context window
1M
Max output
128K
Input price
$5 / M
Output price
$25 / M
Blended price
$10 / M
Output speed
33 tokens/s Kagi
Value
#202 of 219
Knowledge cutoff
January 2026
Input
text, image, pdf

Category scores

Each category score combines every public result we have in that category.

  1. Coding 59.6
  2. Agentic & Tool Use 47.9
  3. Reasoning 53.8
  4. Math 66.7
  5. Knowledge 62.6
  6. Multimodal 41.2
  7. Multilingual 57.3
  8. Instruction Following 78.4
  9. Long Context 46.2
  10. Writing & Preference 75.1
CategoryScoreRankResults
Coding59.6#138
Agentic & Tool Use47.9#108
Reasoning53.8#2913
Math66.7#266
Knowledge62.6#235
Multimodal41.2#383
Multilingual57.3#101
Instruction Following78.4#101
Long Context46.2#251
Writing & Preference75.1#85

Strengths and weaknesses

Categories where Claude Opus 4.7 places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Writing & Preference75.1+21.3#8 of 312, top 3%
Instruction Following78.4+7.1#10 of 305, top 4%
Multilingual57.3+9.9#10 of 297, top 4%

Weakest categories

CategoryScorevs medianRank
Multimodal41.2+2.7#38 of 128, top 30%
Long Context46.2+5.2#25 of 296, top 9%
Reasoning53.8+30.2#29 of 350, top 9%

Closest competitors

The models ranked just above and below Claude Opus 4.7. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Kimi K3#1559.5$6—Compare
GPT-5.4#1659.4$5.6312Compare
GPT-5.6 Terra#1759.2$4.5011Compare
GPT-5.4 Pro#1858.9$67.50—Compare
Claude Opus 4.6#2058.2$1019Compare
Grok 4.6#2156.9$3—Compare
Qwen3.8 Max#2256.8$3—Compare
Gemini 3.1 Pro Preview#2356.7$4.50—Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
SWE-bench Verified83.5%Best of 32maxEpoch AI2026-04-20
FrontierCode38.5%#23 of 37, top 63%Epoch AI
LMArena WebDev1558#31 of 113, top 28%highLMArena2026-10-08
SciCode54.5%#27 of 121, top 23%maxEpoch AI
GSO44.1%#6 of 31, top 20%Epoch AI
GSO44.1%highEpoch AI
WeirdML76.4%Epoch AI
WeirdML76.4%#14 of 119, top 12%highEpoch AI
WeirdML75.5%maxEpoch AI
LMArena Coding1518#8 of 294, top 3%highLMArena2026-10-08
MirrorCode31.1%#5 of 9, top 56%highEpoch AI2026-08-12
ALE-Bench1,323#22 of 105, top 21%Epoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
Terminal-Bench80.2%#3 of 41, top 8%Epoch AI
APEX-Agents49.2%#26 of 49, top 54%maxEpoch AI
OSWorld 2.018.2%#4 of 9, top 45%maxEpoch AI
τ²-bench Banking40.2%#7 of 26, top 27%maxτ²-bench2026-05-05
PostTrainBench28.6%#7 of 11, top 64%xhighEpoch AI
ExploitBench26.5%#3 of 9, top 34%Epoch AI
GBAEval43.8%#12 of 23, top 53%Epoch AI
GDP.pdf21%#19 of 36, top 53%maxEpoch AI
LMArena Search1233#4 of 32, top 13%LMArena2026-08-24
Vending-Bench 210,937#5 of 60, top 9%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
ARC-AGI-268.3%highEpoch AI
ARC-AGI-262.1%lowEpoch AI
ARC-AGI-275.8%#17 of 83, top 21%maxEpoch AI
ARC-AGI-267.5%mediumEpoch AI
SimpleBench61.7%#19 of 77, top 25%Epoch AI
Kagi LLM Benchmark80.7%#6 of 99, top 7%Kagi LLM Benchmark
Kagi LLM Benchmark73.3%Kagi LLM Benchmark
NYT Connections (extended)39%#68 of 91, top 75%high reasoningLech Mazur benchmarks
ARC-AGI-193.5%#20 of 83, top 25%highEpoch AI
ARC-AGI-191%lowEpoch AI
ARC-AGI-192%maxEpoch AI
ARC-AGI-191%mediumEpoch AI
CritPt12%#43 of 134, top 33%maxEpoch AI
Chess Puzzles20%lowEpoch AI2026-07-14
Chess Puzzles7%maxEpoch AI2026-08-06
Chess Puzzles30%#35 of 129, top 28%xhighEpoch AI2026-04-20
Thematic Generalization72.8%#5 of 23, top 22%high reasoningLech Mazur benchmarks
EBR-Bench19%#14 of 24, top 59%maxEpoch AI2026-06-30
LMArena Hard Prompts1506#9 of 297, top 4%highLMArena2026-10-08
Mystery Game Puzzles13%Epoch AI2026-08-06
Mystery Game Puzzles28%#30 of 74, top 41%maxEpoch AI2026-07-25
DTBench94.7%#21 of 151, top 14%maxEpoch AI
LMCA52.2%#18 of 125, top 15%maxEpoch AI
Epoch Capabilities Index156.25#25 of 213, top 12%Epoch AI2026-04-16
ForecastBench60.3#27 of 72, top 38%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
FrontierMath (Tiers 1-3)70.2%#24 of 81, top 30%maxEpoch AI2026-06-10
FrontierMath Tier 431.7%#26 of 63, top 42%maxEpoch AI2026-06-10
MathArena Final-Answer Competitions73.6%#10 of 29, top 35%xhighMathArena
OTIS Mock AIME 2024-202586.7%maxEpoch AI2026-08-06
OTIS Mock AIME 2024-202597.8%#24 of 173, top 14%xhighEpoch AI2026-04-17
ProofBench54%#25 of 77, top 33%maxEpoch AI
LMArena Math1499#14 of 285, top 5%highLMArena2026-10-08
FrontierMath (Feb 2025 set)43.8%#6 of 68, top 9%xhighEpoch AI2026-04-17
FrontierMath Tier 4 (v1)22.9%#8 of 55, top 15%xhighEpoch AI2026-04-17

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond86.4%maxEpoch AI2026-08-06
GPQA Diamond90.2%#37 of 186, top 20%xhighEpoch AI2026-04-17
Humanity's Last Exam36.2%#9 of 41, top 22%Epoch AI
SimpleQA Verified51.7%#23 of 77, top 30%xhighEpoch AI2026-08-27
Vectara Hallucination Rate (lower is better)12%#73 of 96, top 77%Vectara Hallucination Leaderboard
LMArena Expert1521#11 of 273, top 5%LMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1316#5 of 122, top 5%LMArena2026-10-09
Blueprint-Bench 224.5%#22 of 31, top 71%Epoch AI
Furniture Assembly33.3%#21 of 31, top 68%maxEpoch AI2026-09-10
LMArena Document1495#5 of 38, top 14%LMArena2026-09-13

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1480#10 of 297, top 4%highLMArena2026-10-08
LMArena Chinese1531#12 of 285, top 5%highLMArena2026-10-08
LMArena French1503#10 of 223, top 5%highLMArena2026-10-08
LMArena German1495#10 of 231, top 5%highLMArena2026-10-08
LMArena Japanese1472#17 of 211, top 9%highLMArena2026-10-08
LMArena Korean1464#8 of 213, top 4%LMArena2026-10-08
LMArena Russian1494#10 of 283, top 4%highLMArena2026-10-08
LMArena Spanish1495#9 of 226, top 4%highLMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1498#7 of 298, top 3%highLMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1505#8 of 291, top 3%highLMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1490#9 of 297, top 4%highLMArena2026-10-08
LMArena Creative Writing1486#9 of 295, top 4%highLMArena2026-10-08
EQ-Bench Creative Writing1914#13 of 115, top 12%EQ-Bench
EQ-Bench 41311#5 of 28, top 18%EQ-Bench
LMArena Multi-Turn1505#3 of 295, top 2%highLMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
anthropic$5$25$0.502026-10-10
azure$5$25$0.502026-10-10
bedrock$5$25$0.502026-10-10
openrouter$5$25$0.502026-10-10
vertex$5$25$0.502026-10-10

All Anthropic API prices → Estimate your cost →

Compare Claude Opus 4.7

Other Anthropic models

Frequently asked questions

How good is Claude Opus 4.7?

Claude Opus 4.7 by Anthropic ranks 19th of 354 ranked models on the Noometry Index as of October 2026, with a score of 58.3. Its strongest category is writing & preference, where it ranks 8th. API pricing starts at $5 per million input tokens and $25 per million output tokens, with a 1M-token context window.

How much does Claude Opus 4.7 cost?

Claude Opus 4.7 costs $5 per million input tokens and $25 per million output tokens on Anthropic's own API, with cached input at $0.50.

What is Claude Opus 4.7's context window?

Claude Opus 4.7 accepts up to 1M tokens of input and can write up to 128K tokens in one response.

Is Claude Opus 4.7 open source?

No. Claude Opus 4.7 is proprietary and available only through Anthropic's API and partner platforms.

How fast is Claude Opus 4.7?

Claude Opus 4.7 generated about 33 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

What are Claude Opus 4.7's strengths and weaknesses?

Relative to other ranked models, Claude Opus 4.7 places best in writing & preference, instruction following, multilingual and lowest in multimodal, long context, reasoning.

What is Claude Opus 4.7 best at?

Its best category is writing & preference, where it ranks 8th on Noometry.