What does Lmca measure?

Lmca is a knowledge benchmark in the BenchGecko catalog. 110 AI models have been tested on it. Scores range from 3.3 to 80.3 out of 100.

Which model leads on Lmca?

Claude Opus 5.5 from Anthropic leads Lmca with a score of 80.3. The median score across 110 tested models is 41.0.

No · the top score is 80.3 out of 100 (80%). There is still meaningful room for improvement on Lmca.

Does Lmca predict performance on other benchmarks?

Yes · Lmca scores correlate 0.99 with Exploitbench across 5 shared models. Models that do well on Lmca tend to do well on Exploitbench.

How often is Lmca data refreshed?

BenchGecko pulls updates daily. New model scores on Lmca appear as soon as they are published by Epoch AI or the model provider.

Benchmark · KnowledgeSettled

Lmca

Name: Lmca Benchmark
Creator: BenchGecko
License: https://creativecommons.org/licenses/by/4.0/

Updated 2026-09-22

Models tested

110

Top score

80.3

Claude Opus 5.5

Median

41.0

min 3.3

Top-5 spread

σ 2.7

Competitive

The Frontier

Best score over time · one chart, every benchmark

Chart type

Frontier on Lmca rose from 20.6 to 80.3 in 22 months · +59.7 points · latest leader Claude Opus 5.5 from Anthropic.

Pink dots = frontier records · 13 totalClick to open model page

Full rankings

110 models tested · sorted by score

#	Model	Score	Price
1	Claude Opus 5.5· Anthropic	80.3	$4.00
2	Claude Fable 5.1· Anthropic	77.0	$10.00
3	Claude Opus 5· Anthropic	75.8	$5.00
4	GPT-6 Astra· OpenAI	75.7	$10.00
5	Claude Fable 5· Anthropic	71.9	$10.00
6	GPT-5.6 Sol· OpenAI	69.7	$2.00
7	Claude Opus 4.8· Anthropic	67.7	$5.00
8	Claude Opus 4.6· Anthropic	65.6	$5.00
9	GLM 5.3· z-ai	65.3	$1.40
10	GPT-5.6 Terra· OpenAI	64.7	$2.00
11	Gemini 3.1 Pro Preview· Google DeepMind	63.3	$2.00
12	Gemini 3.8 Flash· Google DeepMind	62.2	$0.75
13	Kimi K3· moonshotai	62.0	$2.70
14	Claude Opus 4.7· Anthropic	61.4	$5.00
15	GPT-5.4· OpenAI	61.1	$2.50
16	Gemini 3.7 Flash· Google DeepMind	59.3	$0.75
17	Claude Sonnet 5· Anthropic	58.9	$2.00
18	GPT-5.6 Luna· OpenAI	57.1	$0.20
19	Grok 4.6· xAI	57.0	$2.00
20	Gemini 3.5 Flash· Google DeepMind	55.5	$1.50
21	DeepSeek V4.1 Flash· DeepSeek	55.3	$0.15
22	Claude Sonnet 4.6· Anthropic	54.7	$3.00
23	Qwen3.8 Max (0902)· Alibaba Qwen	54.4	$2.00
24	GLM 5.2· z-ai	53.9	$1.00
25	DeepSeek V4 Pro 0813· DeepSeek	53.5	$0.66
26	Grok 4.5· xAI	53.2	$2.00
27	Gemini 3.6 Flash· Google DeepMind	52.9	$0.75
28	Claude Opus 4.5· Anthropic	52.3	$5.00
29	Qwen3.7 Max· Alibaba Qwen	51.8	$1.25
30	GPT-5.2· OpenAI	51.7	$1.75
31	GPT-5.1· OpenAI	51.6	$1.25
32	Gemini 3 Flash Preview· Google DeepMind	50.7	$0.50
33	Qwen3.6 Max Preview· Alibaba Qwen	50.0	$1.03
34	DeepSeek V4 Flash 0731· DeepSeek	49.1	$0.01
35	Qwen3.8 27B· Alibaba Qwen	48.7	$0.42
36	DeepSeek V4 Pro· DeepSeek	48.5	$0.21
37	GPT-5.4 Mini· OpenAI	48.0	$0.75
38	GPT-5· OpenAI	47.0	$1.25
39	o3· OpenAI	46.7	$2.00
40	Gemma 4 31B· Google DeepMind	46.2	$0.09
41	Claude Sonnet 4.5· Anthropic	45.6	$3.00
42	Grok 4.20· xAI	45.5	$1.25
43	o3 Pro· OpenAI	45.3	$20.00
44	Grok 4.3· xAI	45.1	$1.25
45	Qwen3.5 397B A17B· Alibaba Qwen	44.6	$0.55
46	Gemini 3.5 Flash Lite· Google DeepMind	44.2	$0.30
47	Qwen3.7 Plus· Alibaba Qwen	44.2	$0.32
48	Claude Opus 4· Anthropic	44.0	$15.00
49	Kimi K2.6· moonshotai	43.8	$0.95
50	Claude Opus 4.1· Anthropic	43.6	$15.00
51	Nemotron 3 Ultra· NVIDIA	43.4	$0.50
52	GPT-5.4 Nano· OpenAI	43.4	$0.20
53	Qwen 3.5 Plus (hosted 397B-A17B)· Alibaba Qwen	42.8	—
54	DeepSeek V4 Flash· DeepSeek	42.2	$0.03
55	Gemini 3.1 Flash Lite· Google DeepMind	41.2	$0.25
56	Gemini 2.5 Pro· Google DeepMind	40.9	$1.25
57	Qwen3.6 27B· Alibaba Qwen	40.6	$0.32
58	GPT-5 Mini· OpenAI	40.2	$0.25
59	MiniMax M3· minimax	39.6	$0.30
60	Qwen3.6 Plus· Alibaba Qwen	38.9	$0.33
61	Qwen3.6 Flash· Alibaba Qwen	36.5	$0.19
62	Claude Haiku 4.5· Anthropic	36.4	$1.00
63	Qwen3.6 35B A3B· Alibaba Qwen	34.9	$0.15
64	Gemma 4 26B A4B · Google DeepMind	34.9	$0.07
65	Qwen3.5-35B-A3B· Alibaba Qwen	34.7	$0.15
66	Qwen3 235B A22B Thinking 2507· Alibaba Qwen	34.5	$0.23
67	Qwen3.5-Flash· Alibaba Qwen	34.3	$0.07
68	DeepSeek V3.2 Exp· DeepSeek	34.3	$0.27
69	Claude Sonnet 4· Anthropic	34.1	$3.00
70	DeepSeek V3.2· DeepSeek	33.9	$0.28
71	Qwen3 Max· Alibaba Qwen	33.3	$0.78
72	Gemini 2.5 Flash· Google DeepMind	32.3	$0.30
73	o4 Mini· OpenAI	31.2	$1.10
74	Mistral Medium 3.5· Mistral AI	30.7	$1.50
75	GPT-4.1· OpenAI	30.1	$2.00
76	Qwen3 235B A22B· Alibaba Qwen	29.4	$0.46
77	Qwen3.5-9B· Alibaba Qwen	28.8	$0.10
78	DeepSeek V3.1· DeepSeek	28.6	$0.25
79	Qwen3 235B A22B Instruct 2507· Alibaba Qwen	27.7	$0.09
80	o1· OpenAI	26.2	$15.00
81	gpt-oss-120b· OpenAI	26.1	$0.04
82	GPT-4.1 Mini· OpenAI	24.9	$0.40
83	o3 Mini· OpenAI	22.3	$1.10
84	Qwen3 14B· Alibaba Qwen	21.4	$0.12
85	Gemini 2.5 Flash Lite· Google DeepMind	21.3	$0.10
86	Llama 3.3 70B Instruct· Meta	20.6	$0.10
87	Qwen3 32B· Alibaba Qwen	20.3	$0.08
88	GPT-4 Turbo· OpenAI	20.1	$10.00
89	Claude 3 Opus· Anthropic	20.0	—
90	GPT-4o (2024-11-20)· OpenAI	19.5	$2.50
91	Llama 4 Maverick· Meta	18.7	$0.19
92	Qwen3 30B A3B· Alibaba Qwen	18.6	$0.12
93	DeepSeek V3· DeepSeek	18.2	$0.26
94	Llama 3.1 70B Instruct· Meta	17.5	$0.40
95	gpt-oss-20b· OpenAI	17.1	$0.02
96	Qwen2.5 72B Instruct· Alibaba Qwen	15.8	$0.36
97	Gemma 3 27B· Google DeepMind	14.4	$0.08
98	Llama 4 Scout· Meta	14.1	$0.10
99	GPT-4o-mini· OpenAI	12.2	$0.15
100	GPT-3.5 Turbo (older v0613)· OpenAI	11.3	$1.00
101	Claude 3 Haiku· Anthropic	10.4	$0.25
102	Qwen3 8B· Alibaba Qwen	10.4	$0.12
103	GPT-5 Nano· OpenAI	9.3	$0.05
104	Gemma 2 27B· Google DeepMind	8.3	$0.65
105	Qwen2.5 7B Instruct· Alibaba Qwen	7.5	$0.10
106	GPT-4.1 Nano· OpenAI	6.4	$0.10
107	Llama 3.1 8B Instruct· Meta	6.3	$0.02
108	Command R (08-2024)· Cohere	5.9	$0.15
109	Gemma 3 12B· Google DeepMind	5.3	$0.05
110	Gemma 3 4B· Google DeepMind	3.3	$0.05