BenchLeader

LMCA

A capability assessment with a published score ceiling.

As of 19 Sept 2026, Claude Opus 5 leads LMCA on BenchLeader with 63.3%, ahead of Claude Fable 5 at 60.3%, across 119 model configurations with a published result.

Published by
Epoch AI Benchmarking Hub
Category
Reasoning
Index weight
Reference only
Models
119
Data as of
19 Sept 2026

CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.

What the test looks like

A broad assessment scored 0 to 100 with a ceiling of 85 for the current version.

How it is scored

Score, as published.

What to keep in mind

The ceiling means top scores compress; new.

119 of 119
#
1Claude Opus 5maxAnthropic63.3%69.92026-07-24
2Claude Fable 5maxAnthropic60.3%65.82026-06-09
3GPT-5.6 SolOpenAI59.2%2026-07-09
4GPT-5.6 SolmaxOpenAI58.4%68.82026-07-09
5Claude Opus 4.8maxAnthropic57.5%64.22026-05-28
6Claude Opus 4.6maxAnthropic55.8%59.22026-02-05
7GPT-5.5xhighOpenAI54.3%67.62026-04-23
8GPT-5.5 ProxhighOpenAI53.9%67.12026-04-23
9Gemini 3.1 ProhighGoogle53.8%60.72026-02-19
10Kimi K3maxMoonshot AIopen ↗52.7%67.22026-07-16
11GPT-5.6 TerramaxOpenAI52.2%65.02026-07-09
12Claude Opus 4.7maxAnthropic52.2%64.32026-04-16
13GPT-5.4xhighOpenAI52.0%65.42026-03-05
14Gemini 3.7 FlashhighGoogle50.4%64.62026-08-13
15Muse Spark 1.1highMeta49.9%2026-07-09
16Claude Sonnet 5maxAnthropic49.3%60.52026-06-30
17GPT-5.6 LunamaxOpenAI48.5%60.22026-07-09
18Grok 4.6xhighxAI48.5%64.62026-08-12
19Muse Spark 1.2xhighMeta48.4%64.12026-08-05
20Gemini 3.5 FlashhighGoogle47.1%63.62026-05-19
21Claude Sonnet 4.6maxAnthropic46.5%57.62026-02-17
22Qwen3 8xhighAlibaba46.2%57.72026-08-02
23GLM-5.2maxZhipu AIopen ↗45.8%63.92026-06-16
24Grok 4.5highxAI45.2%61.02026-07-08
25Gemini 3.6 FlashhighGoogle44.9%61.72026-07-21
26Claude Opus 4.5highAnthropic44.5%57.72025-11-24
27Qwen3 7maxAlibaba44.0%61.02026-05-19
28GPT-5.2xhighOpenAI43.9%62.02025-12-11
29GPT-5.1highOpenAI43.9%58.72025-11-13
30Gemini 3 FlashhighGoogle43.1%58.72025-12-17
31Qwen3 6maxAlibaba42.5%63.32026-04-20
32DeepSeek V4 Flash 0731maxDeepSeekopen ↗41.7%54.42026-07-31
33DeepSeek V4 PromaxDeepSeekopen ↗41.2%64.22026-04-24
34GPT-5.4 minixhighOpenAI40.8%55.72026-03-17
35GPT-5highOpenAI40.0%58.62025-08-07
36o3highOpenAI39.7%53.02025-04-16
37Gemma 4 31BGoogleopen ↗39.3%52.72026-04-02
38Claude Sonnet 4.5Anthropic38.8%54.32025-09-29
39Grok 4.20xAI38.7%2026-02-17
40o3-prohighOpenAI38.5%54.52025-06-10
41Grok 4.3highxAI38.3%58.22026-04-17
42Qwen3.5 397B-A17BAlibabaopen ↗37.9%56.92026-02-13
43Gemini 3.5 Flash LitehighGoogle37.6%44.32026-07-21
44Qwen3.7 PlusAlibaba37.6%59.92026-06-02
45Claude Opus 4Anthropic37.4%52.92025-05-22
46Kimi K2.6Moonshot AIopen ↗37.3%60.52026-04-20
47Claude Opus 4.1Anthropic37.1%51.92025-08-05
48Nemotron 3 UltraNVIDIA36.9%48.32026-06-04
49GPT-5.4 nanoxhighOpenAI36.9%54.62026-03-17
50Qwen3.5 PlusAlibaba36.4%2026-02-16
51DeepSeek V4 FlashmaxDeepSeekopen ↗35.9%62.12026-04-24
52Gemini 3.1 Flash LitehighGoogle35.0%49.12026-03-03
53Gemini 2.5 ProGoogle34.8%54.52025-06-17
54Qwen3.6 27BAlibaba34.5%44.72026-04-22
55GPT-5 minihighOpenAI34.2%55.32025-08-07
56Qwen3.5 27BAlibaba34.0%53.02026-02-24
57MiniMax-M3MiniMaxopen ↗33.7%57.52026-06-01
58Qwen3.6 PlusAlibabaopen33.1%58.52026-03-31
59Qwen3.5 122B-A10Bno reasoningAlibaba32.2%48.12026-02-24
60Qwen3.6 FlashAlibaba31.0%44.52026-04-27

Cite as: BenchLeader, “LMCA leaderboard”, https://www.benchleader.com/benchmarks/lmca, data as of 19 Sept 2026.

LMCA: questions

What does LMCA measure?
A broad assessment scored 0 to 100 with a ceiling of 85 for the current version. Scores are reported in percent of tasks solved; higher is better.
Which AI model leads LMCA?
Claude Opus 5 leads LMCA with 63.3% as of 19 Sept 2026, ahead of Claude Fable 5 at 60.3%.
How many models have LMCA results?
119 model configurations have a LMCA result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
Who runs LMCA and how often is it updated?
LMCA is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results every morning and records the date each result was published.
Does LMCA count toward the BenchLeader Index?
No. LMCA is shown for reference but left out of the composite index.