Code Arena

Compare the performance of AI models on agentic coding tasks involving multi-step reasoning and tool use

Last Updated

Feb 1, 2026

Total Votes

125,171

Total Models

/

	Rank Spread
1	1◄─►1	claude-opus-4-5-20251101-thinking-32k	1500	+9/-9	8,297	Anthropic	Proprietary
2	2◄─►5	gpt-5.2-high	1472	+16/-16	1,689	OpenAI	Proprietary
3	2◄─►4	claude-opus-4-5-20251101	1470	+9/-9	8,488	Anthropic	Proprietary
4	3◄─►7	gemini-3-pro	1453	+8/-8	14,466	Google	Proprietary
5	2◄─►7	kimi-k2.5-thinking	1447	+16/-16	1,488	Moonshot	Modified MIT
6	4◄─►7	gemini-3-flash	1443	+8/-8	9,926	Google	Proprietary
7	4◄─►7	glm-4.7	1441	+10/-10	5,117	Z.ai	MIT
8	8◄─►11	minimax-m2.1-preview	1409	+9/-9	7,470	MiniMax	MIT
9	8◄─►15	gemini-3-flash (thinking-minimal)	1401	+9/-9	5,941	Google	Proprietary
10	8◄─►16	gpt-5.2	1397	+15/-15	1,633	OpenAI	Proprietary
11	8◄─►16	gpt-5-medium	1394	+12/-12	3,926	OpenAI	Proprietary
12	9◄─►16	claude-opus-4-1-20250805	1389	+8/-8	8,979	Anthropic	Proprietary
13	9◄─►16	gpt-5.1-medium	1389	+9/-9	6,433	OpenAI	Proprietary
14	9◄─►16	claude-sonnet-4-5-20250929-thinking-32k	1387	+8/-8	11,664	Anthropic	Proprietary
15	9◄─►16	claude-sonnet-4-5-20250929	1386	+7/-7	13,262	Anthropic	Proprietary
16	10◄─►16	deepseek-v3.2-thinking	1377	+11/-11	3,995	DeepSeek	MIT
17	17◄─►19	glm-4.6	1356	+8/-8	8,736	Z.ai	MIT
18	17◄─►20	gpt-5.1	1350	+8/-8	10,624	OpenAI	Proprietary
19	17◄─►22	mimo-v2-flash (non-thinking)	1347	+10/-10	4,518	Xiaomi	MIT
20	18◄─►22	gpt-5.2-codex	1333	+12/-12	3,235	OpenAI	Proprietary
21	19◄─►22	kimi-k2-thinking-turbo	1330	+8/-8	10,173	Moonshot	Modified MIT
22	19◄─►23	gpt-5.1-codex	1329	+9/-9	6,500	OpenAI	Proprietary
23	22◄─►25	minimax-m2	1312	+9/-9	8,841	MiniMax	Apache 2.0
24	23◄─►26	deepseek-v3.2	1301	+10/-10	5,081	DeepSeek	MIT
25	23◄─►27	claude-haiku-4-5-20251001	1297	+8/-8	11,416	Anthropic	Proprietary
26	24◄─►27	deepseek-v3.2-exp	1287	+10/-10	5,133	DeepSeek	MIT
27	25◄─►27	qwen3-coder-480b-a35b-instruct	1283	+7/-7	11,168	Alibaba	Apache 2.0
28	28◄─►30	KAT-Coder-Pro-V1	1259	+15/-15	1,956	KwaiKAT	Proprietary
29	28◄─►31	gpt-5.1-codex-mini	1243	+17/-17	1,536	OpenAI	Proprietary
30	28◄─►31	grok-4-1-fast-reasoning	1238	+10/-10	5,824	xAI	Proprietary
31	29◄─►33	mistral-large-3	1223	+20/-20	1,039	Mistral	Apache 2.0
32	31◄─►33	gemini-2.5-pro	1206	+13/-13	3,454	Google	Proprietary
33	31◄─►33	grok-4.1-thinking	1205	+19/-19	1,266	xAI	Proprietary
34	34◄─►35	grok-4-fast-reasoning	1153	+22/-22	969	xAI	Proprietary
35	34◄─►36	grok-code-fast-1	1141	+21/-21	1,016	xAI	Proprietary
36	35◄─►36	devstral-medium-2507	1098	+22/-22	1,020	Mistral	Proprietary

Code Arena

Remove Style Control Leaderboard Plots

Fraction of Model A Wins for All Non-tied A vs. B Battles

Battle Count for Each Combination of Models (without Ties)

Confidence Intervals on Model Strength (via Bootstrapping)

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)