Reasoning benchmark

LMCA leaderboard

As of October 2026, Claude Opus 5.5 has the highest published LMCA score on Noometry at 68.2%, out of 125 models with results.

Last verified October 10, 2026

About LMCA

A description with primary sources is being prepared for this benchmark.

Category: Reasoning
Introduced: 2026
Unit: Percent (random guessing ≈ 0%)
Official site: epoch.ai

Top 15 models

Claude Opus 5.5 68.2%
Claude Fable 5.1 65.5%
Claude Opus 5 64.5%
GPT-6 Astra 64.4%
Claude Fable 5 61.1%
GPT-5.6 Sol 59.2%
GPT-6 Sol 59.1%
Claude Opus 4.8 57.5%
Claude Opus 4.6 55.8%
GLM-5.3 55.5%
GPT-5.6 Terra 55%
GPT-5.5 54.3%
GPT-5.5 Pro 53.9%
Muse Spark 1.3 53.9%
Gemini 3.1 Pro Preview 53.8%
5055606570

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

All results

LMCA results by model
#	Model	Provider	Score	Setting	Source
1	Claude Opus 5.5	Anthropic	68.2%	max	Epoch AI
2	Claude Fable 5.1	Anthropic	65.5%		Epoch AI
3	Claude Opus 5	Anthropic	64.5%	xhigh	Epoch AI
4	GPT-6 Astra	OpenAI	64.4%	xhigh	Epoch AI
5	Claude Fable 5	Anthropic	61.1%	high	Epoch AI
6	GPT-5.6 Sol	OpenAI	59.2%	promax	Epoch AI
7	GPT-6 Sol	OpenAI	59.1%	max	Epoch AI
8	Claude Opus 4.8	Anthropic	57.5%	max	Epoch AI
9	Claude Opus 4.6	Anthropic	55.8%	max	Epoch AI
10	GLM-5.3	Z.ai (Zhipu)	55.5%		Epoch AI
11	GPT-5.6 Terra	OpenAI	55%	max	Epoch AI
12	GPT-5.5	OpenAI	54.3%	xhigh	Epoch AI
13	GPT-5.5 Pro	OpenAI	53.9%	xhigh	Epoch AI
14	Muse Spark 1.3	Meta	53.9%		Epoch AI
15	Gemini 3.1 Pro Preview	Google	53.8%	high	Epoch AI
16	Gemini 3.8 Flash	Google	52.9%	medium	Epoch AI
17	Kimi K3	Moonshot AI	52.7%	max	Epoch AI
18	Claude Opus 4.7	Anthropic	52.2%	max	Epoch AI
19	GPT-5.4	OpenAI	52%	xhigh	Epoch AI
20	Gemini 3.7 Flash	Google	50.4%	high	Epoch AI
21	Claude Sonnet 5	Anthropic	50%	high	Epoch AI
22	Muse Spark 1.1	Meta	49.9%	high	Epoch AI
23	Grok 4.7	xAI	49.4%	xhigh	Epoch AI
24	GPT-5.6 Luna	OpenAI	48.5%	max	Epoch AI
25	Grok 4.6	xAI	48.5%	xhigh	Epoch AI
26	Muse Spark 1.2	Meta	48.4%	xhigh	Epoch AI
27	Gemini 3.5 Flash	Google	47.1%	high	Epoch AI
28	DeepSeek V4.1 Flash	DeepSeek	47%		Epoch AI
29	Claude Sonnet 4.6	Anthropic	46.5%	max	Epoch AI
30	Qwen3.8 Max	Alibaba (Qwen)	46.2%	xhigh	Epoch AI
31	GLM-5.2	Z.ai (Zhipu)	45.8%	max	Epoch AI
32	DeepSeek V4 Pro	DeepSeek	45.5%		Epoch AI
33	Grok 4.5	xAI	45.2%	high	Epoch AI
34	Gemini 3.6 Flash	Google	44.9%	high	Epoch AI
35	GPT-6 Luna	OpenAI	44.5%	max	Epoch AI
36	Claude Opus 4.5	Anthropic	44.5%	high	Epoch AI
37	Qwen3.7 Max	Alibaba (Qwen)	44%		Epoch AI
38	GPT-5.2	OpenAI	43.9%	xhigh	Epoch AI
39	GPT-5.1	OpenAI	43.9%	high	Epoch AI
40	Gemini 3 Flash Preview	Google	43.1%	high	Epoch AI
41	Qwen3.6 Max Preview	Alibaba (Qwen)	42.5%		Epoch AI
42	DeepSeek V4 Flash	DeepSeek	41.7%	max	Epoch AI
43	Qwen3.8 27B	Alibaba (Qwen)	41.4%		Epoch AI
44	GPT-5.4 mini	OpenAI	40.8%	xhigh	Epoch AI
45	GPT-5	OpenAI	40%	high	Epoch AI
46	o3	OpenAI	39.7%	high	Epoch AI
47	Gemma 4 31B IT	Google	39.3%		Epoch AI
48	Claude Sonnet 4.5	Anthropic	38.8%		Epoch AI
49	Grok 4.20 (Non-Reasoning)	xAI	38.7%		Epoch AI
50	o3-pro	OpenAI	38.5%	high	Epoch AI
51	Grok 4.3	xAI	38.3%	high	Epoch AI
52	Qwen3.5 397B-A17B	Alibaba (Qwen)	37.9%		Epoch AI
53	Gemini 3.5 Flash Lite	Google	37.6%	high	Epoch AI
54	Inkling	Thinking Machines Lab	37.6%	xhigh	Epoch AI
55	Qwen3.7 Plus	Alibaba (Qwen)	37.6%		Epoch AI
56	Claude Opus 4	Anthropic	37.4%		Epoch AI
57	Kimi K2.6	Moonshot AI	37.3%		Epoch AI
58	Claude Opus 4.1	Anthropic	37.1%		Epoch AI
59	Nemotron 3 Ultra	NVIDIA	36.9%		Epoch AI
60	GPT-5.4 nano	OpenAI	36.9%	xhigh	Epoch AI
61	Qwen3.5 Plus	Alibaba (Qwen)	36.4%		Epoch AI
62	Gemini 3.1 Flash Lite	Google	35%	high	Epoch AI
63	Gemini 2.5 Pro	Google	34.8%		Epoch AI
64	Qwen3.6 27B	Alibaba (Qwen)	34.5%		Epoch AI
65	GPT-5 Mini	OpenAI	34.2%	high	Epoch AI
66	Qwen3.5 27B	Alibaba (Qwen)	34%		Epoch AI
67	MiniMax-M3	MiniMax	33.7%		Epoch AI
68	Qwen3.6 Plus	Alibaba (Qwen)	33.1%		Epoch AI
69	Qwen3.5 122B-A10B	Alibaba (Qwen)	32.2%	none	Epoch AI
70	Qwen3.6 Flash	Alibaba (Qwen)	31%		Epoch AI
71	Claude Haiku 4.5	Anthropic	30.9%		Epoch AI
72	Qwen3.6 35B-A3B	Alibaba (Qwen)	29.7%		Epoch AI
73	Gemma 4 26B A4B IT	Google	29.7%		Epoch AI
74	MiMo-V2.5-Pro	Xiaomi	29.5%		Epoch AI
75	Qwen3.5 35B-A3B	Alibaba (Qwen)	29.5%		Epoch AI
76	Qwen3 235B-A22B	Alibaba (Qwen)	29.3%		Epoch AI
77	Qwen3.5-Flash	Alibaba (Qwen)	29.1%		Epoch AI
78	DeepSeek-V3.2-Exp	DeepSeek	29.1%	thinking	Epoch AI
79	Claude Sonnet 4	Anthropic	29%		Epoch AI
80	DeepSeek-V3.1-Terminus	DeepSeek	28.6%		Epoch AI
81	Qwen3 Max	Alibaba (Qwen)	28.3%		Epoch AI
82	Gemini 2.5 Flash	Google	27.5%		Epoch AI
83	o4-mini	OpenAI	26.5%	high	Epoch AI
84	Mistral Medium	Mistral AI	26.1%		Epoch AI
85	GPT-4.1	OpenAI	25.6%		Epoch AI
86	Qwen3.5-9B	Alibaba (Qwen)	24.5%		Epoch AI
87	DeepSeek-V3.1	DeepSeek	24.3%	thinking	Epoch AI
88	Qwen Plus	Alibaba (Qwen)	24%		Epoch AI
89	Qwen3-30B-A3B	Alibaba (Qwen)	22.4%		Epoch AI
90	o1	OpenAI	22.3%	high	Epoch AI
91	gpt-oss-120b	OpenAI	22.1%	high	Epoch AI
92	GPT-4.1 mini	OpenAI	21.1%		Epoch AI
93	Mistral Small	Mistral AI	20.6%		Epoch AI
94	o3-mini	OpenAI	19%	high	Epoch AI
95	Qwen3 14B	Alibaba (Qwen)	18.2%		Epoch AI
96	Gemini 2.5 Flash-Lite	Google	18.1%		Epoch AI
97	Llama-3.3-70B-Instruct	Meta	17.5%		Epoch AI
98	Qwen3 32B	Alibaba (Qwen)	17.3%		Epoch AI
99	GPT-4	OpenAI	17.1%		Epoch AI
100	Claude 3 Opus	Anthropic	17%		Epoch AI
101	Mistral Large	Mistral AI	16.7%		Epoch AI
102	GPT-4o	OpenAI	16.6%		Epoch AI
103	Llama 4 Maverick	Meta	15.9%		Epoch AI
104	DeepSeek-V3	DeepSeek	15.5%		Epoch AI
105	Llama 3.1-70B	Meta	14.8%		Epoch AI
106	gpt-oss-20b	OpenAI	14.5%	high	Epoch AI
107	Qwen2.5 72B Instruct	Alibaba (Qwen)	13.4%		Epoch AI
108	Gemma 3 27B	Google	12.3%		Epoch AI
109	Llama 4 Scout	Meta	12%		Epoch AI
110	GPT-4o mini	OpenAI	10.4%		Epoch AI
111	Command A	Cohere	10.3%		Epoch AI
112	GPT-4 Turbo	OpenAI	9.8%		Epoch AI
113	GPT-3.5-turbo	OpenAI	9.7%		Epoch AI
114	Command R	Cohere	9.2%		Epoch AI
115	Claude 3 Haiku	Anthropic	8.8%		Epoch AI
116	Qwen3 8B	Alibaba (Qwen)	8.8%		Epoch AI
117	GPT-5 Nano	OpenAI	7.9%	high	Epoch AI
118	Gemma 2 27B	Google	7.1%		Epoch AI
119	Qwen2.5 7B Instruct	Alibaba (Qwen)	6.4%		Epoch AI
120	Ministral 3B	Mistral AI	5.5%		Epoch AI
121	GPT-4.1 nano	OpenAI	5.5%		Epoch AI
122	Llama 3.1-8B	Meta	5.4%		Epoch AI
123	Command R+	Cohere	5%		Epoch AI
124	Gemma 3 12B	Google	4.5%		Epoch AI
125	Gemma 3 4B	Google	2.8%		Epoch AI

Compare the leaders

Other reasoning benchmarks

Frequently asked questions

Which model has the highest LMCA score?

As of October 2026, Claude Opus 5.5 has the highest published LMCA score on Noometry at 68.2%, out of 125 models with results.

What is the best open-weight model on LMCA?

GLM-5.3 has the highest LMCA accuracy among open-weight models at 55.5%, ranking 10 of 125 overall.