Arena Elo

Elo ratings from blind pairwise votes: a human is shown two anonymous outputs for the same prompt and picks the better one. The votes are collected and published by Arena — we ingest the ratings, we do not run the votes and we do not adjust them.

Text Image Video Speech

Arena Elo

Human preference rating from blind pairwise votes, ingested from Arena

Text ArenaText Arena (style controlled)

Text Arena

Elo from blind pairwise preference votes, collected and published by Arena. Higher is better. We ingest these ratings; we do not run the votes and we do not run our own evaluations.

⇩

The Known Good

Text Arena — full ranking

406 models rated 11,989,921 votes countedRatings captured 4 Sep 2026 Source: Arena

#	Model	Creator	Elo	95% CI	Votes	Known Good Index
1	claude-opus-5-max	Anthropic	1,505	±6	16,839	—
2	claude-opus-5-high	Anthropic	1,504	±5	34,617	—
3	claude-opus-4-6-high	Anthropic	1,503	±4	72,110	—
3	claude-opus-4-6-thinking	Anthropic	1,503	±4	72,425	—
4	Anthropic: Claude Opus 4.6	Anthropic	1,498	±3	76,032	89
5	Anthropic: Claude Fable 5	Anthropic	1,494	±5	26,977	—
6	gemini-3.7-flash-high	Google	1,491	±8	5,685	—
7	claude-opus-4-7-thinking	Anthropic	1,490	±4	60,222	—
7	claude-opus-4-7-high	Anthropic	1,490	±4	60,123	—
8	Meta: Muse Spark 1.2	Meta	1,488	±10	3,244	—
9	gemini-3.5-flash-high	Google	1,484	±4	33,874	—
10	Anthropic: Claude Opus 4.7	Anthropic	1,483	±4	61,256	92
11	Google: Gemini 3.1 Pro Preview	Google	1,480	±3	102,763	97
12	Qwen: Qwen3.8 Max	Alibaba	1,480	±6	13,136	—
13	Gemini 3 Pro	Google	1,480	±4	40,675	—
13	Google: Gemini 3.6 Flash	Google	1,479	±6	13,559	—
14	Meta: Muse Spark 1.1	Meta	1,479	±5	23,768	—
15	gemini-3.6-flash-high	Google	1,476	±5	22,009	—
16	kimi-k3-max	Moonshot AI	1,476	±5	17,895	—
17	gemini-3.5-flash-medium	Google	1,476	±5	32,274	—
18	glm-5.3-max	Z.ai	1,474	±7	7,401	—
19	qwen3.7-max-preview	Alibaba	1,474	±10	3,710	—
20	muse-spark	Meta	1,474	±6	13,571	—
13	MoonshotAI: Kimi K3	Moonshot AI	1,473	±10	3,619	—
21	gpt-5.5-high	OpenAI	1,472	±4	63,265	—
22	qwen3.5-max-preview	Alibaba	1,471	±5	21,487	—
23	Z.ai: GLM 5.3 Flash	Z.ai	1,471	±9	4,399	—
24	gpt-5.4-high	OpenAI	1,470	±4	60,657	—
25	ernie-5.1	Baidu	1,468	±5	37,137	—
26	glm-5.2-max	Z.ai	1,467	±5	33,758	—
27	OpenAI: GPT-5.5	OpenAI	1,467	±4	64,686	93
29	Mi Xiaomi: MiMo-V2.5-Pro	Xiaomi	1,466	±4	57,416	—
22	Z.ai: GLM 5.2	Z.ai	1,464	±6	18,017	91
30	Z.ai: GLM 5.1	Z.ai	1,463	±4	45,403	90
31	claude-opus-4-8-high	Anthropic	1,461	±4	48,467	—
28	claude-opus-4-8-thinking	Anthropic	1,460	±5	40,409	—
32	Anthropic: Claude Sonnet 4.6	Anthropic	1,458	±4	66,343	80
33	Google: Gemini 2.5 Pro	Google	1,458	±2	122,575	65
34	MoonshotAI: Kimi K2.6	Moonshot AI	1,455	±5	37,573	92
35	gpt-5.6-sol-xhigh	OpenAI	1,455	±5	23,153	—
36	Qwen: Qwen3.7 Plus	Alibaba	1,454	±4	37,854	—
37	OpenAI: GPT-5.4	OpenAI	1,453	±4	63,615	91
38	Anthropic: Claude Opus 4.8	Anthropic	1,452	±4	49,123	—
39	SpaceXAI: Grok 4.5	xAI	1,452	±5	25,873	—
41	DeepSeek: DeepSeek V4 Pro 0423	DeepSeek	1,451	±4	54,243	93
40	grok-4.20-beta-0309-reasoning	xAI	1,451	±4	62,269	—
42	Anthropic: Claude Opus 4.5	Anthropic	1,450	±3	70,057	72
43	grok-4.20-multi-agent-beta-0309	xAI	1,450	±4	60,845	—
44	dola-seed-2.0-pro	ByteDance	1,448	±3	74,291	—
45	Az amazon-nova-experimental-chat-26-02-10	Amazon	1,448	±10	3,373	—
46	claude-opus-4-5-20251101-high-32k	Anthropic	1,447	±4	36,271	—
43	claude-opus-4-5-20251101-thinking-32k	Anthropic	1,447	±4	37,213	—
47	gpt-5.6-terra-xhigh	OpenAI	1,447	±5	23,942	—
48	Z.ai: GLM 5	Z.ai	1,446	±4	27,643	77
49	Qwen: Qwen3.6 Max Preview	Alibaba	1,446	±8	5,192	89
50	kimi-k2.5-thinking	Moonshot AI	1,446	±3	70,620	—
51	deepseek-v4-pro-high-preview	DeepSeek	1,445	±4	51,599	—
52	nvidia-nemotron-3-ultra-550b-a55b-nvfp4	NVIDIA	1,445	±7	10,704	—
47	deepseek-v4-pro-thinking	DeepSeek	1,445	±4	44,669	—
53	grok-4.20-beta1	xAI	1,444	±5	26,629	—
54	ernie-5.0-0110	Baidu	1,444	±4	34,883	—
55	grok-4.6-high	xAI	1,444	±10	3,453	—
56	claude-sonnet-5-high	Anthropic	1,443	±5	31,096	—
57	ernie-5.0-preview-1203	Baidu	1,443	±7	9,555	—
58	gemini-3-flash	Google	1,442	±3	85,926	—
59	gpt-5.1-high	OpenAI	1,442	±4	40,296	—
60	Google: Gemma 4 31B	Google	1,442	±8	5,900	—
61	Z.ai: GLM 4.6	Z.ai	1,440	±4	35,061	—
62	deepseek-v4-pro-high-20260813	DeepSeek	1,440	±8	5,388	—
63	Tencent: Hy3	Tencent	1,440	±8	6,911	—
64	qwen3-max-preview	Alibaba	1,439	±5	27,182	—
65	gpt-5.2-chat-latest-20260210	OpenAI	1,439	±4	34,218	—
66	claude-sonnet-4-5-20250929	Anthropic	1,438	±3	79,668	66
67	TM Thinking Machines: Inkling	Thinking Machines	1,438	±5	21,973	—
68	Qwen: Qwen3.5 397B A17B	Alibaba	1,438	±3	73,400	—
69	Qwen: Qwen3.8 27B	Alibaba	1,437	±8	6,626	—
70	grok-4.1-thinking	xAI	1,437	±3	64,192	—
71	Qwen: Qwen3.6 Plus	Alibaba	1,437	±4	45,374	79
72	Z.ai: GLM 5V Turbo	Z.ai	1,437	±7	9,369	—
73	Mi mimo-v2-pro	Xiaomi	1,436	±5	24,388	—
74	Google: Gemini 3.5 Flash Lite	Google	1,436	±5	21,843	—
75	grok-4.1	xAI	1,436	±3	66,380	—
76	Z.ai: GLM 4.7	Z.ai	1,436	±6	11,891	69
77	MiniMax: MiniMax M3	MiniMax	1,435	±4	44,544	—
78	Google: Gemma 4 26B A4B	Google	1,435	±8	5,814	—
79	claude-sonnet-4-5-20250929-high-32k	Anthropic	1,434	±3	81,031	—
73	claude-sonnet-4-5-20250929-thinking-32k	Anthropic	1,433	±3	82,549	—
80	DeepSeek: DeepSeek V4 Flash 0423	DeepSeek	1,432	±4	48,963	—
81	ernie-5.0-preview-1022	Baidu	1,430	±9	4,662	—
82	gpt-5.6-luna-xhigh	OpenAI	1,430	±5	24,433	—
83	Z.ai: GLM 4.5	Z.ai	1,430	±5	23,706	—
84	chatgpt-4o-latest-20250326	OpenAI	1,429	±3	80,693	—
85	mistral-large-3	Mistral	1,428	±3	65,336	—
86	Mi Xiaomi: MiMo-V2.5	Xiaomi	1,428	±4	44,540	—
87	DeepSeek: R1 0528	DeepSeek	1,427	±6	18,087	—
88	Me longcat-flash-chat-2602-exp	meituan	1,426	±5	27,991	—
89	grok-3-preview-02-24	xAI	1,425	±4	32,409	—
90	deepseek-v3.2-exp-thinking	DeepSeek	1,425	±7	8,911	—
91	DeepSeek: DeepSeek V3.2	DeepSeek	1,425	±4	46,487	—
92	Mistral: Mistral Medium 3.1	Mistral	1,425	±3	92,735	—
93	deepseek-v4-flash-high-preview	DeepSeek	1,424	±4	48,705	—
85	deepseek-v4-flash-thinking	DeepSeek	1,423	±4	46,445	—
94	Me longcat-flash-chat	meituan	1,423	±6	11,167	—
95	DeepSeek: DeepSeek V3.2 Exp	DeepSeek	1,423	±6	11,758	—
96	OpenAI: GPT-5.1	OpenAI	1,422	±4	43,002	74
97	Mi mimo-v2-omni	Xiaomi	1,422	±6	19,448	—
98	Mistral: Mistral Medium 3.5	Mistral	1,421	±7	11,017	—
99	Az amazon-nova-experimental-chat-12-10	Amazon	1,421	±10	3,633	—
100	Qwen: Qwen3 VL 235B A22B Instruct	Alibaba	1,420	±7	11,228	—
101	kimi-k2.5-instant	Moonshot AI	1,420	±7	7,996	—
102	deepseek-v3.2-thinking	DeepSeek	1,420	±4	40,510	—
103	Qwen: Qwen3 235B A22B Instruct 2507	Alibaba	1,420	±3	95,685	—
104	deepseek-v3.1-terminus-thinking	DeepSeek	1,420	±10	3,371	—
105	gpt-5.5-instant	OpenAI	1,420	±5	25,860	—
106	DeepSeek: DeepSeek V3.1	DeepSeek	1,420	±6	14,601	—
107	claude-opus-4-1-20250805-thinking-16k	Anthropic	1,419	±3	48,799	—
108	Qwen: Qwen3.5-122B-A10B	Alibaba	1,418	±4	28,376	—
109	Anthropic: Claude Opus 4.1	Anthropic	1,418	±3	75,882	50
110	Qwen: Qwen3 Next 80B A3B Instruct	Alibaba	1,417	±5	22,520	—
111	Google: Gemini 2.5 Flash	Google	1,417	±2	122,770	—
112	gpt-4.5-preview-2025-02-27	OpenAI	1,417	±6	14,547	—
113	DeepSeek: DeepSeek V3.1 Terminus	DeepSeek	1,417	±10	3,614	—
114	gpt-5.2-high	OpenAI	1,417	±4	47,566	—
115	deepseek-v3.1-thinking	DeepSeek	1,417	±7	11,460	—
116	Az amazon-nova-experimental-chat-11-10	Amazon	1,416	±4	25,099	—
117	Google: Gemini 3.1 Flash Lite Preview	Google	1,415	±4	60,505	—
118	Qwen: Qwen3 235B A22B Thinking 2507	Alibaba	1,415	±7	8,796	—
119	kimi-k2-thinking-turbo	Moonshot AI	1,414	±3	61,124	—
120	TM Thinking Machines: Inkling Small	Thinking Machines	1,414	±6	14,940	—
121	qwen3-max-2025-09-23	Alibaba	1,413	±6	8,939	—
122	OpenAI: GPT-5.2	OpenAI	1,412	±3	79,073	81
123	gpt-5.4-mini-high	OpenAI	1,412	±4	59,472	—
124	grok-4-0709	xAI	1,411	±4	39,285	68
125	Mi mimo-v2-flash (non-thinking)	Xiaomi	1,411	±4	46,161	—
126	o3-2025-04-16	OpenAI	1,410	±4	58,561	—
127	Qwen: Qwen3.5-27B	Alibaba	1,408	±4	27,246	—
128	grok-4-1-fast-reasoning	xAI	1,408	±3	55,403	—
129	gemini-2.5-flash-preview-09-2025	Google	1,407	±4	32,350	—
130	grok-4-fast-chat	xAI	1,407	±8	6,336	—
131	hunyuan-vision-1.5-thinking	Tencent	1,406	±12	2,170	—
132	gpt-5-high	OpenAI	1,406	±4	31,418	—
133	hunyuan-hy3-preview	Tencent	1,405	±8	6,613	—
134	MiniMax: MiniMax M2.7	MiniMax	1,405	±4	65,411	—
135	St StepFun: Step 3.5 Flash	stepfun	1,404	±4	57,225	—
136	OpenAI: GPT-5 Chat	OpenAI	1,404	±4	30,964	—
137	hunyuan-t1-20250711	Tencent	1,401	±9	4,575	—
138	Qwen: Qwen3 VL 235B A22B Thinking	Alibaba	1,401	±7	7,789	—
139	grok-4-fast-reasoning	xAI	1,399	±5	17,965	—
140	Az amazon-nova-experimental-chat-26-01-10	Amazon	1,398	±10	3,349	—
141	SpaceXAI: Grok 4.3	xAI	1,398	±4	65,024	—
142	Qwen: Qwen3.5-Flash	Alibaba	1,398	±4	58,299	—
143	Qwen: Qwen3.5-35B-A3B	Alibaba	1,396	±4	29,067	—
144	claude-haiku-4-5-20251001	Anthropic	1,395	±3	124,979	67
145	Az amazon-nova-experimental-chat-10-20	Amazon	1,395	±6	11,362	—
146	Mi mimo-v2-flash (thinking)	Xiaomi	1,394	±6	10,751	—
147	qwen3-235b-a22b-no-thinking	Alibaba	1,394	±5	37,377	—
148	muse-glimmer	Meta	1,392	±10	3,696	—
149	minimax-m2.1-preview	MiniMax	1,391	±5	16,638	—
150	OpenAI: GPT-5.3 Chat	OpenAI	1,388	±4	32,749	—
151	Z.ai: GLM 4.5 Air	Z.ai	1,384	±4	30,368	—
152	Qwen: Qwen3 30B A3B Instruct 2507	Alibaba	1,384	±5	23,222	—
153	gpt-4.1-2025-04-14	OpenAI	1,383	±4	49,913	—
154	gemini-2.5-flash-lite-preview-09-2025-no-thinking	Google	1,379	±3	46,384	—
155	MoonshotAI: Kimi K2 0905	Moonshot AI	1,379	±6	11,564	—
156	nvidia-nemotron-3-super-120b-a12b	NVIDIA	1,378	±7	7,461	—
157	solar-pro4	Upstage	1,377	±12	2,437	—
158	claude-opus-4-20250514-thinking-16k	Anthropic	1,377	±4	35,725	—
159	Z.ai: GLM 4.6V	Z.ai	1,376	±11	2,760	—
160	hunyuan-turbos-20250416	Tencent	1,376	±6	10,583	—
161	DeepSeek: DeepSeek V3 0324	DeepSeek	1,375	±4	44,786	64
162	gpt-5-mini-high	OpenAI	1,373	±5	26,551	—
163	gpt-5.4-nano-high	OpenAI	1,373	±4	58,562	—
164	DeepSeek: R1	DeepSeek	1,372	±5	18,524	75
165	MoonshotAI: Kimi K2 0711	Moonshot AI	1,371	±5	27,023	—
166	Mistral: Mistral Medium 3	Mistral	1,370	±5	32,660	—
167	grok-3-mini-high	xAI	1,369	±5	15,535	—
168	gemini-2.5-flash-lite-preview-06-17-thinking	Google	1,369	±4	32,209	—
169	Qwen: Qwen3 Next 80B A3B Thinking	Alibaba	1,368	±6	13,389	—
170	qwen2.5-max	Alibaba	1,366	±4	32,417	—
171	Qwen: Qwen3 235B A22B	Alibaba	1,366	±5	25,865	—
172	OpenAI: gpt-oss-120b	OpenAI	1,366	±4	29,935	62
173	claude-opus-4-20250514	Anthropic	1,366	±4	42,841	68
174	o1-2024-12-17	OpenAI	1,366	±4	27,807	—
175	Az amazon-nova-experimental-chat-10-09	Amazon	1,364	±11	2,776	—
176	grok-3-mini-beta	xAI	1,364	±5	20,838	—
177	granite-4.2-30b	IBM	1,363	±11	3,287	—
178	Az Amazon: Nova 2 Lite	Amazon	1,363	±6	12,117	—
179	AG ling-flash-2.0	ant-group	1,362	±7	6,852	—
180	MiniMax: MiniMax M2.5	MiniMax	1,359	±4	40,895	—
181	In Inception: Mercury 2	inception	1,358	±11	3,043	—
182	Google: Gemma 3 27B	Google	1,358	±4	46,434	47
183	Qwen3-Coder-480B-A35B-Instruct	Alibaba	1,356	±5	25,081	—
184	I3 intellect-3	Unknown	1,356	±8	5,282	—
185	GD gemini-2.0-flash-001	Google DeepMind,Google	1,354	±4	43,346	61
186	o4-mini-2025-04-16	OpenAI	1,353	±4	44,618	—
187	o1-preview	OpenAI	1,353	±5	31,122	—
188	Z.ai: GLM 4.7 Flash	Z.ai	1,353	±6	11,513	—
189	St step-3	stepfun	1,351	±7	6,280	—
190	claude-sonnet-4-20250514-thinking-32k	Anthropic	1,350	±5	33,955	—
191	nvidia-nemotron-3-nano-30b-a3b-bf16	NVIDIA	1,348	±5	15,338	—
192	MiniMax: MiniMax M1	MiniMax	1,342	±4	34,028	—
193	AA Arcee AI: Trinity Large Thinking	arcee-ai	1,342	±5	28,999	—
194	MiniMax: MiniMax M2	MiniMax	1,342	±8	6,748	—
195	gpt-4.1-mini-2025-04-14	OpenAI	1,340	±4	38,630	—
196	Qwen: Qwen3 32B	Alibaba	1,340	±9	3,926	—
197	claude-sonnet-4-20250514	Anthropic	1,339	±4	38,942	—
198	TL trinity-large-preview	Unknown	1,339	±4	29,770	—
199	mistral-small-2506	Mistral	1,338	±5	17,401	—
200	nvidia-llama-3.3-nemotron-super-49b-v1.5	NVIDIA	1,338	±10	3,316	—
201	OpenAI: o3 Mini High	OpenAI	1,336	±5	18,589	—
202	St step-1o-turbo-202506	stepfun	1,336	±7	8,856	—
203	Google: Gemma 3 12B	Google	1,334	±9	3,829	—
204	Z.ai: GLM 4.5V	Z.ai	1,333	±8	4,818	—
205	DeepSeek: DeepSeek V3	DeepSeek	1,332	±5	21,770	62
206	AG ring-flash-2.0	ant-group	1,332	±7	6,904	—
207	nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4	NVIDIA	1,332	±9	4,533	—
208	Cohere: Command A	Cohere	1,331	±3	55,434	—
209	glm-4-plus-0111	Z.ai	1,331	±8	5,760	—
210	gemini-2.0-flash-lite-preview-02-05	Google	1,329	±4	24,955	—
211	QwQ-32B	Alibaba	1,329	±4	25,094	69
212	qwen-plus-0125	Alibaba	1,326	±8	5,819	—
213	St step-2-16k-exp-202412	stepfun	1,321	±9	4,833	—
214	gpt-5-nano-high	OpenAI	1,320	±7	8,128	—
215	hunyuan-turbos-20250226	Tencent	1,320	±12	2,220	—
216	llama-3.1-nemotron-ultra-253b-v1	NVIDIA	1,319	±12	2,549	—
217	gemini-1.5-pro-002	Google	1,319	±3	55,606	—
218	OpenAI: o3 Mini	OpenAI	1,319	±4	56,653	86
219	IBM: Granite 4.2 8B	IBM	1,318	±11	3,119	—
220	o1-mini	OpenAI	1,317	±4	51,981	65
221	Qwen: Qwen3 30B A3B	Alibaba	1,317	±5	26,093	—
222	claude-3-7-sonnet-20250219-thinking-32k	Anthropic	1,314	±4	38,088	—
223	A2 olmo-3.1-32b-instruct	Allen Institute for AI	1,312	±6	11,463	—
224	hunyuan-turbo-0110	Tencent	1,311	±11	2,290	—
225	llama-3.3-nemotron-49b-super-v1	NVIDIA	1,308	±12	2,218	—
226	Google: Gemma 3n 4B	Google	1,305	±5	22,084	—
227	grok-2-2024-08-13	xAI	1,304	±4	63,498	—
228	YL yi-lightning	Unknown	1,301	±5	27,332	—
229	OpenAI: GPT-4o (2024-05-13)	OpenAI	1,300	±3	112,881	—
230	A2 AllenAI: Olmo 3 32B Think	Allen Institute for AI	1,300	±8	5,682	—
231	claude-3-7-sonnet-20250219	Anthropic	1,299	±4	42,452	74
232	qwen2.5-plus-1127	Alibaba	1,299	±6	10,187	—
233	granite-4.2-3b	IBM	1,299	±12	3,097	—
234	claude-3-5-sonnet-20241022	Anthropic	1,298	±3	87,693	45
235	deepseek-v2.5-1210	DeepSeek	1,294	±8	6,795	—
236	A2 molmo-2-8b	Allen Institute for AI	1,294	±21	792	—
237	IBM: Granite 4.1 8B	IBM	1,292	±10	4,027	—
238	AV athene-v2-chat	Unknown	1,291	±5	24,739	—
239	Google: Gemma 3 4B	Google	1,291	±9	4,171	—
240	glm-4-plus	Z.ai	1,289	±5	26,126	—
241	hunyuan-large-2025-02-10	Tencent	1,288	±10	3,738	—
242	OpenAI: gpt-oss-20b	OpenAI	1,287	±6	10,386	42
243	llama-4-maverick-17b-128e-instruct	Meta	1,287	±4	39,349	—
244	gemini-1.5-flash-002	Google	1,286	±4	34,902	—
245	OpenAI: GPT-4o-mini (2024-07-18)	OpenAI	1,286	±4	68,698	—
246	gpt-4.1-nano-2025-04-14	OpenAI	1,285	±8	6,103	—
247	IA mercury	inception-ai	1,284	±14	1,929	—
248	llama-3.1-405b-instruct-bf16	Meta	1,283	±4	41,375	—
249	llama-3.1-nemotron-70b-instruct	NVIDIA	1,282	±8	7,140	—
250	OpenAI: GPT-4o (2024-08-06)	OpenAI	1,282	±4	45,499	—
251	qwen-max-0919	Alibaba	1,282	±6	16,478	—
252	llama-3.1-405b-instruct-fp8	Meta	1,282	±4	59,656	—
253	claude-3-5-sonnet-20240620	Anthropic	1,281	±3	82,419	—
254	grok-2-mini-2024-08-13	xAI	1,280	±4	52,567	—
255	Llama-4-Scout-17B-16E-Instruct	Meta	1,279	±5	29,739	—
256	gemini-advanced-0514	Google	1,278	±5	50,148	—
257	mistral-small-3.1-24b-instruct-2503	Mistral	1,277	±5	32,621	—
258	Meta: Llama 3.3 70B Instruct	Meta	1,274	±3	54,411	34
259	hunyuan-standard-2025-02-10	Tencent	1,274	±10	3,904	—
260	gemini-1.5-pro-001	Google	1,273	±4	79,138	—
261	gpt-4-turbo-2024-04-09	OpenAI	1,271	±4	98,114	—
262	deepseek-v2.5	DeepSeek	1,271	±5	24,572	—
263	A2 olmo-3.1-32b-think	Allen Institute for AI	1,271	±7	8,112	—
264	Qwen2.5 72B Instruct	Alibaba	1,269	±4	39,406	—
265	Mistral Large 2407	Mistral	1,266	±4	45,459	—
266	mistral-large-2411	Mistral	1,265	±4	28,073	37
267	A7 athene-70b-0725	Unknown	1,265	±6	19,621	—
268	gpt-4-1106-preview	OpenAI	1,263	±4	100,105	—
269	hunyuan-large-vision	Tencent	1,263	±9	5,260	—
270	gpt-4-0125-preview	OpenAI	1,262	±4	93,439	—
271	claude-3-opus-20240229	Anthropic	1,262	±3	194,909	32
272	Meta: Llama 3.1 70B Instruct	Meta	1,261	±4	55,240	—
273	Az amazon.nova-pro-v1:0	Amazon	1,259	±5	24,745	—
274	A2 llama-3.1-tulu-3-70b	Allen Institute for AI	1,256	±10	2,846	—
275	claude-3-5-haiku-20241022	Anthropic	1,255	±3	68,972	30
276	magistral-medium-2506	Mistral	1,255	±6	11,472	—
277	Re reka-core-20240904	Reka AI	1,248	±7	7,312	—
278	ibm-granite-h-small	IBM	1,240	±9	5,534	—
279	gemini-1.5-flash-001	Google	1,239	±4	62,833	—
280	J1 jamba-1.5-large	Unknown	1,237	±7	8,662	—
281	Mistral: Mistral Small 3	Mistral	1,233	±6	14,681	—
282	Google: Gemma 2 27B	Google	1,231	±3	75,754	21
283	Qwen2.5 Coder 32B Instruct	Alibaba	1,230	±8	5,432	—
284	Cohere: Command R+ (08-2024)	Cohere	1,229	±7	9,866	—
285	Az amazon.nova-lite-v1:0	Amazon	1,228	±5	19,372	—
286	llama-3.1-nemotron-51b-instruct	NVIDIA	1,228	±10	3,749	—
287	gemma-2-9b-it-simpo	Google	1,227	±7	10,072	—
288	glm-4-0520	Z.ai	1,226	±7	9,788	—
289	gemini-1.5-flash-8b-001	Google	1,226	±4	35,558	—
290	nemotron-4-340b-instruct	NVIDIA	1,225	±5	19,659	—
291	c4ai-aya-expanse-32b	Cohere	1,224	±5	27,124	—
292	llama-3-70b-instruct	Meta	1,221	±4	156,876	—
293	claude-3-sonnet-20240229	Anthropic	1,218	±4	109,284	—
294	A2 olmo-2-0325-32b-instruct	Allen Institute for AI	1,218	±11	3,334	—
295	Re reka-flash-20240904	Reka AI	1,218	±7	7,536	—
296	Microsoft: Phi 4	Microsoft	1,216	±5	24,126	41
297	Az amazon.nova-micro-v1:0	Amazon	1,208	±5	19,364	—
298	gemma-2-9b-it	Google	1,207	±4	54,611	13
299	gpt-4-0314	OpenAI	1,206	±5	54,173	—
300	command-r-plus	Cohere	1,204	±4	77,554	—
301	qwen2-72b-instruct	Alibaba	1,203	±5	37,325	—
302	hunyuan-standard-256k	Tencent	1,202	±12	2,728	—
303	Anthropic: Claude 3 Haiku	Anthropic	1,194	±4	117,701	—
304	A2 llama-3.1-tulu-3-8b	Allen Institute for AI	1,193	±11	2,896	—
305	deepseek-coder-v2	DeepSeek	1,191	±6	15,147	—
306	ministral-8b-2410	Mistral	1,191	±9	4,781	—
308	J1 jamba-1.5-mini	Unknown	1,186	±7	8,858	—
309	Meta: Llama 3.1 8B Instruct	Meta	1,186	±4	49,605	—
310	gpt-4-0613	OpenAI	1,186	±4	88,723	—
311	c4ai-aya-expanse-8b	Cohere	1,185	±7	9,818	—
312	mistral-large-2402	Mistral	1,176	±5	62,436	—
313	qwen1.5-110b-chat	Alibaba	1,174	±6	26,195	—
314	0A Yi-1.5-34B-Chat	01.AI	1,173	±5	24,146	—
315	Re reka-flash-21b-20240226-online	Reka AI	1,170	±7	15,450	—
316	qwen1.5-72b-chat	Alibaba	1,166	±5	39,302	—
317	llama-3-8b-instruct	Meta	1,165	±4	104,642	—
318	mistral-medium	Mistral	1,165	±5	34,550	—
319	Re reka-flash-21b-20240226	Reka AI	1,165	±6	24,806	—
320	Cohere: Command R (08-2024)	Cohere	1,163	±5	54,036	—
321	mixtral-8x22b-instruct-v0.1	Mistral	1,162	±5	51,416	—
322	QwQ-32B-Preview	Alibaba	1,162	±11	3,231	—
323	I5 internlm2_5-20b-chat	Unknown	1,158	±7	9,901	—
324	gemma-2-2b-it	Google	1,156	±4	46,616	—
325	granite-3.1-8b-instruct	IBM	1,149	±11	3,090	—
326	gemini-pro-dev-api	Google	1,149	±7	18,354	—
327	ZO zephyr-orpo-141b-A35b-v0.1	Unknown	1,143	±11	4,652	—
328	phi-3-medium-4k-instruct	Microsoft	1,137	±5	25,055	—
329	qwen1.5-32b-chat	Alibaba	1,137	±6	21,741	—
330	SL starling-lm-7b-beta	Unknown	1,132	±7	16,056	—
331	Mixtral-8x7B-Instruct-v0.1	Mistral	1,131	±4	73,503	—
332	gemini-pro	Google	1,130	±11	6,390	—
333	0A Yi-34B-Chat	01.AI	1,128	±7	15,483	—
334	qwen1.5-14b-chat	Alibaba	1,127	±7	17,839	—
335	granite-3.1-2b-instruct	IBM	1,127	±11	3,188	—
336	gpt-3.5-turbo-0125	OpenAI	1,125	±5	66,207	—
337	T2 tulu-2-dpo-70b	Unknown	1,121	±10	6,535	—
338	wizardlm-70b	Microsoft	1,119	±9	8,214	—
339	DI dbrx-instruct-preview	Unknown	1,119	±6	32,191	—
340	llama-2-70b-chat	Meta	1,115	±5	38,492	—
341	nous-hermes-2-mixtral-8x7b-dpo	Nous Research	1,112	±12	3,777	—
342	Phi-3-small-8k-instruct	Microsoft	1,110	±6	17,766	—
343	Meta: Llama 3.2 3B Instruct	Meta	1,109	±8	7,936	—
344	SL starling-lm-7b-alpha	Unknown	1,107	±8	10,224	—
345	O3 openchat-3.5-0106	Unknown	1,106	±8	12,637	—
346	V3 vicuna-33b	Unknown	1,105	±6	22,479	—
347	deepseek-llm-67b-chat	DeepSeek	1,105	±12	4,932	—
348	snowflake-arctic-instruct	Snowflake	1,101	±6	32,832	—
349	llama2-70b-steerlm-chat	NVIDIA	1,098	±12	3,585	—
350	O3 openchat-3.5	Unknown	1,097	±10	7,968	—
351	granite-3.0-8b-instruct	IBM	1,096	±9	6,638	—
352	gpt-3.5-turbo-1106	OpenAI	1,094	±9	16,619	—
353	gemma-1.1-7b-it	Google	1,094	±6	23,893	—
354	O2 openhermes-2.5-mistral-7b	Unknown	1,092	±10	5,006	—
355	mistral-7b-instruct-v0.2	Mistral	1,089	±7	19,402	—
356	llama-2-13b-chat	Meta	1,084	±7	19,174	—
357	qwen1.5-7b-chat	Alibaba	1,083	±10	4,737	—
358	solar-10.7b-instruct-v1.0	Upstage	1,083	±13	4,155	—
359	D2 dolphin-2.2.1-mistral-7b	Unknown	1,081	±15	1,679	—
360	phi-3-mini-4k-instruct-june-2024	Microsoft	1,080	±6	12,297	—
361	granite-3.0-2b-instruct	IBM	1,080	±8	6,837	—
362	wizardlm-13b	Microsoft	1,076	±9	7,044	—
363	Phi-3-mini-4k-instruct	Microsoft	1,073	±6	20,118	—
364	Z7 zephyr-7b-beta	Unknown	1,069	±9	11,118	—
365	M3 mpt-30b-chat	Unknown	1,068	±12	2,572	—
366	codellama-34b-instruct	Meta	1,065	±9	7,366	—
367	Z7 zephyr-7b-alpha	Unknown	1,058	±16	1,785	—
368	V1 vicuna-13b	Unknown	1,057	±7	19,367	—
369	codellama-70b-instruct	Meta	1,057	±18	1,143	—
370	gemma-7b-it	Google	1,056	±10	8,925	—
371	Meta: Llama 3.2 1B Instruct	Meta	1,054	±8	8,045	—
372	F1 falcon-180b-chat	Unknown	1,054	±17	1,295	—
373	llama-2-7b-chat	Meta	1,053	±7	14,148	—
374	G3 guanaco-33b	Unknown	1,053	±12	2,921	—
375	qwen-14b-chat	Alibaba	1,051	±11	4,964	—
376	phi-3-mini-128k-instruct	Microsoft	1,050	±7	20,685	—
377	S1 smollm2-1.7b-instruct	Unknown	1,042	±14	2,199	—
378	SN stripedhyena-nous-7b	Unknown	1,038	±11	5,182	—
379	A2 olmo-7b-instruct	Allen Institute for AI	1,032	±11	6,328	—
380	V7 vicuna-7b	Unknown	1,031	±9	6,923	—
381	palm-2	Google	1,027	±9	8,554	—
382	mistral-7b-instruct	Mistral	1,023	±9	8,977	—
383	gemma-1.1-2b-it	Google	1,022	±8	10,854	—
384	gemma-2b-it	Google	1,002	±11	4,780	—
385	qwen1.5-4b-chat	Alibaba	997	±9	7,597	—
386	K1 koala-13b	Unknown	989	±10	6,965	—
387	C6 chatglm3-6b	Unknown	972	±12	4,658	—
388	G1 gpt4all-13b-snoozy	Unknown	956	±15	1,743	—
389	M7 mpt-7b-chat	Unknown	955	±12	3,924	—
390	R4 RWKV-4-Raven-14B	Unknown	948	±11	4,845	—
391	C6 chatglm2-6b	Unknown	938	±13	2,658	—
392	A1 alpaca-13b	Unknown	933	±11	5,745	—
393	chatglm-6b	Z.ai	918	±12	4,914	—
394	OP oasst-pythia-12b	Unknown	916	±11	6,310	—
395	FT fastchat-t5-3b	Unknown	894	±12	4,203	—
396	ST stablelm-tuned-alpha-7b	Unknown	866	±13	3,287	—
397	DV dolly-v2-12b	Unknown	851	±13	3,412	—
398	llama-13b	Meta	833	±16	2,391	—

The Known Good

Text Arena (style controlled) — full ranking

406 models rated 11,989,921 votes countedRatings captured 4 Sep 2026 Source: Arena

#	Model	Creator	Elo	95% CI	Votes	Known Good Index
1	Anthropic: Claude Fable 5	Anthropic	1,507	±5	26,977	—
2	claude-opus-4-6-high	Anthropic	1,505	±4	72,110	—
2	claude-opus-4-6-thinking	Anthropic	1,505	±4	72,425	—
3	claude-opus-4-7-thinking	Anthropic	1,502	±4	60,222	—
3	claude-opus-4-7-high	Anthropic	1,502	±4	60,123	—
4	Meta: Muse Spark 1.2	Meta	1,499	±10	3,244	—
5	Anthropic: Claude Opus 4.6	Anthropic	1,498	±3	76,032	89
6	Anthropic: Claude Opus 4.7	Anthropic	1,494	±4	61,256	92
7	claude-opus-5-high	Anthropic	1,492	±5	34,617	—
8	Meta: Muse Spark 1.1	Meta	1,492	±5	23,768	—
9	gemini-3.7-flash-high	Google	1,491	±8	5,685	—
10	kimi-k3-max	Moonshot AI	1,489	±5	17,895	—
11	muse-spark	Meta	1,488	±6	13,571	—
12	claude-opus-5-max	Anthropic	1,488	±6	16,839	—
13	Google: Gemini 3.1 Pro Preview	Google	1,487	±3	102,763	97
10	MoonshotAI: Kimi K3	Moonshot AI	1,486	±10	3,619	—
14	Gemini 3 Pro	Google	1,486	±4	40,675	—
15	Google: Gemini 3.6 Flash	Google	1,484	±6	13,559	—
15	glm-5.3-max	Z.ai	1,483	±7	7,401	—
16	gpt-5.6-sol-xhigh	OpenAI	1,483	±5	23,153	—
17	gpt-5.5-high	OpenAI	1,482	±4	63,265	—
18	claude-opus-4-8-high	Anthropic	1,482	±4	48,467	—
17	claude-opus-4-8-thinking	Anthropic	1,481	±5	40,409	—
19	gemini-3.6-flash-high	Google	1,480	±5	22,009	—
20	Qwen: Qwen3.8 Max	Alibaba	1,479	±6	13,136	—
21	gemini-3.5-flash-high	Google	1,479	±4	33,874	—
22	OpenAI: GPT-5.5	OpenAI	1,477	±4	64,686	93
23	gpt-5.4-high	OpenAI	1,476	±4	60,657	—
24	gpt-5.2-chat-latest-20260210	OpenAI	1,476	±4	34,218	—
25	grok-4.20-beta1	xAI	1,475	±5	26,629	—
26	gemini-3.5-flash-medium	Google	1,474	±5	32,274	—
27	Z.ai: GLM 5.3 Flash	Z.ai	1,474	±9	4,399	—
29	qwen3.7-max-preview	Alibaba	1,474	±10	3,710	—
30	gpt-5.5-instant	OpenAI	1,474	±5	25,860	—
31	Anthropic: Claude Opus 4.8	Anthropic	1,473	±4	49,123	—
32	claude-opus-4-5-20251101-high-32k	Anthropic	1,473	±4	36,271	—
29	claude-opus-4-5-20251101-thinking-32k	Anthropic	1,473	±4	37,213	—
33	Anthropic: Claude Sonnet 4.6	Anthropic	1,472	±4	66,343	80
34	glm-5.2-max	Z.ai	1,472	±5	33,758	—
35	grok-4.20-beta-0309-reasoning	xAI	1,472	±4	62,269	—
36	grok-4.20-multi-agent-beta-0309	xAI	1,471	±4	60,845	—
37	SpaceXAI: Grok 4.5	xAI	1,470	±5	25,873	—
38	Anthropic: Claude Opus 4.5	Anthropic	1,469	±3	70,057	72
30	Z.ai: GLM 5.2	Z.ai	1,469	±6	18,017	91
39	Mi Xiaomi: MiMo-V2.5-Pro	Xiaomi	1,468	±4	57,416	—
40	ernie-5.1	Baidu	1,468	±5	37,137	—
41	gpt-5.6-terra-xhigh	OpenAI	1,466	±5	23,942	—
42	Z.ai: GLM 5.1	Z.ai	1,466	±4	45,403	90
43	grok-4.1-thinking	xAI	1,466	±3	64,192	—
44	OpenAI: GPT-5.4	OpenAI	1,466	±4	63,615	91
45	qwen3.5-max-preview	Alibaba	1,466	±5	21,487	—
46	claude-sonnet-5-high	Anthropic	1,462	±5	31,096	—
47	grok-4.6-high	xAI	1,461	±10	3,453	—
48	MoonshotAI: Kimi K2.6	Moonshot AI	1,461	±5	37,573	92
49	Qwen: Qwen3.6 Max Preview	Alibaba	1,460	±8	5,192	89
50	deepseek-v4-pro-high-20260813	DeepSeek	1,460	±8	5,388	—
51	grok-4.1	xAI	1,459	±3	66,380	—
52	gemini-3-flash	Google	1,458	±3	85,926	—
53	Z.ai: GLM 5	Z.ai	1,458	±4	27,643	77
54	DeepSeek: DeepSeek V4 Pro 0423	DeepSeek	1,458	±4	54,243	93
55	Google: Gemini 3.5 Flash Lite	Google	1,457	±5	21,843	—
56	claude-sonnet-4-5-20250929-high-32k	Anthropic	1,456	±3	81,031	—
54	claude-sonnet-4-5-20250929-thinking-32k	Anthropic	1,456	±3	82,549	—
57	dola-seed-2.0-pro	ByteDance	1,456	±3	74,291	—
58	Tencent: Hy3	Tencent	1,455	±8	6,911	—
59	Qwen: Qwen3.7 Plus	Alibaba	1,455	±5	37,854	—
60	claude-sonnet-4-5-20250929	Anthropic	1,455	±3	79,668	66
61	gpt-5.1-high	OpenAI	1,455	±4	40,296	—
62	deepseek-v4-pro-high-preview	DeepSeek	1,455	±4	51,599	—
53	deepseek-v4-pro-thinking	DeepSeek	1,455	±4	44,669	—
63	gpt-5.6-luna-xhigh	OpenAI	1,452	±5	24,433	—
64	Google: Gemma 4 31B	Google	1,451	±8	5,900	—
65	kimi-k2.5-thinking	Moonshot AI	1,451	±3	70,620	—
66	claude-opus-4-1-20250805-thinking-16k	Anthropic	1,450	±3	48,799	—
67	OpenAI: GPT-5.3 Chat	OpenAI	1,449	±4	32,749	—
68	ernie-5.0-preview-1203	Baidu	1,449	±7	9,555	—
69	Mi mimo-v2-pro	Xiaomi	1,448	±5	24,388	—
70	gpt-5.4-mini-high	OpenAI	1,448	±4	59,472	—
71	Anthropic: Claude Opus 4.1	Anthropic	1,448	±3	75,882	50
72	ernie-5.0-0110	Baidu	1,446	±4	34,883	—
73	Google: Gemini 2.5 Pro	Google	1,446	±2	122,575	65
74	gpt-4.5-preview-2025-02-27	OpenAI	1,445	±6	14,547	—
75	Qwen: Qwen3.6 Plus	Alibaba	1,444	±4	45,374	79
76	chatgpt-4o-latest-20250326	OpenAI	1,443	±3	80,693	—
77	MiniMax: MiniMax M3	MiniMax	1,443	±4	44,544	—
78	SpaceXAI: Grok 4.3	xAI	1,443	±4	65,024	—
79	Z.ai: GLM 4.7	Z.ai	1,442	±6	11,891	69
80	Qwen: Qwen3.5 397B A17B	Alibaba	1,441	±3	73,400	—
81	TM Thinking Machines: Inkling	Thinking Machines	1,439	±5	21,973	—
82	OpenAI: GPT-5.1	OpenAI	1,439	±4	43,002	74
83	deepseek-v4-flash-high-preview	DeepSeek	1,438	±4	48,705	—
84	Google: Gemma 4 26B A4B	Google	1,438	±8	5,814	—
85	gpt-5.2-high	OpenAI	1,438	±4	47,566	—
77	deepseek-v4-flash-thinking	DeepSeek	1,438	±4	46,445	—
86	OpenAI: GPT-5.2	OpenAI	1,436	±3	79,073	81
87	DeepSeek: DeepSeek V4 Flash 0423	DeepSeek	1,436	±4	48,963	—
88	Me longcat-flash-chat-2602-exp	meituan	1,436	±5	27,991	—
89	Qwen: Qwen3.8 27B	Alibaba	1,435	±8	6,626	—
90	qwen3-max-preview	Alibaba	1,435	±5	27,182	—
91	gpt-5-high	OpenAI	1,434	±5	31,418	—
92	Mi Xiaomi: MiMo-V2.5	Xiaomi	1,434	±4	44,540	—
93	Z.ai: GLM 5V Turbo	Z.ai	1,433	±7	9,369	—
94	Google: Gemini 3.1 Flash Lite Preview	Google	1,432	±4	60,505	—
95	o3-2025-04-16	OpenAI	1,431	±4	58,561	—
96	kimi-k2.5-instant	Moonshot AI	1,431	±7	7,996	—
97	grok-4-1-fast-reasoning	xAI	1,430	±3	55,403	—
98	kimi-k2-thinking-turbo	Moonshot AI	1,430	±3	61,124	—
99	Mi mimo-v2-omni	Xiaomi	1,430	±6	19,448	—
100	muse-glimmer	Meta	1,427	±10	3,696	—
101	Mistral: Mistral Medium 3.5	Mistral	1,427	±7	11,017	—
102	OpenAI: GPT-5 Chat	OpenAI	1,427	±4	30,964	—
103	nvidia-nemotron-3-ultra-550b-a55b-nvfp4	NVIDIA	1,426	±7	10,704	—
104	claude-opus-4-20250514-thinking-16k	Anthropic	1,426	±4	35,725	—
105	Az amazon-nova-experimental-chat-26-02-10	Amazon	1,426	±10	3,373	—
106	DeepSeek: DeepSeek V3.2	DeepSeek	1,425	±4	46,487	—
107	deepseek-v3.2-exp-thinking	DeepSeek	1,425	±7	8,911	—
108	Z.ai: GLM 4.6	Z.ai	1,425	±4	35,061	—
109	qwen3-max-2025-09-23	Alibaba	1,424	±7	8,939	—
110	Qwen: Qwen3 235B A22B Instruct 2507	Alibaba	1,423	±3	95,685	—
111	deepseek-v3.2-thinking	DeepSeek	1,423	±4	40,510	—
112	DeepSeek: DeepSeek V3.2 Exp	DeepSeek	1,422	±6	11,758	—
113	DeepSeek: R1 0528	DeepSeek	1,421	±6	18,087	—
114	ernie-5.0-preview-1022	Baidu	1,419	±9	4,662	—
115	grok-4-fast-chat	xAI	1,418	±8	6,336	—
116	MoonshotAI: Kimi K2 0905	Moonshot AI	1,418	±7	11,564	—
117	MoonshotAI: Kimi K2 0711	Moonshot AI	1,418	±5	27,023	—
118	DeepSeek: DeepSeek V3.1	DeepSeek	1,417	±6	14,601	—
119	Qwen: Qwen3.5-122B-A10B	Alibaba	1,417	±4	28,376	—
120	deepseek-v3.1-terminus-thinking	DeepSeek	1,417	±10	3,371	—
121	deepseek-v3.1-thinking	DeepSeek	1,416	±7	11,460	—
122	MiniMax: MiniMax M2.7	MiniMax	1,416	±4	65,411	—
123	Az amazon-nova-experimental-chat-26-01-10	Amazon	1,415	±10	3,349	—
124	DeepSeek: DeepSeek V3.1 Terminus	DeepSeek	1,415	±10	3,614	—
125	gpt-4.1-2025-04-14	OpenAI	1,414	±4	49,913	—
126	claude-opus-4-20250514	Anthropic	1,414	±4	42,841	68
127	Qwen: Qwen3 VL 235B A22B Instruct	Alibaba	1,414	±7	11,228	—
128	mistral-large-3	Mistral	1,414	±3	65,336	—
129	claude-haiku-4-5-20251001	Anthropic	1,413	±3	124,979	67
130	hunyuan-hy3-preview	Tencent	1,413	±8	6,613	—
131	grok-3-preview-02-24	xAI	1,411	±4	32,409	—
132	Z.ai: GLM 4.5	Z.ai	1,411	±5	23,706	—
133	grok-4-0709	xAI	1,411	±4	39,285	68
134	Google: Gemini 2.5 Flash	Google	1,410	±2	122,770	—
135	Mistral: Mistral Medium 3.1	Mistral	1,409	±3	92,735	—
136	Qwen: Qwen3.5-27B	Alibaba	1,408	±4	27,246	—
137	TM Thinking Machines: Inkling Small	Thinking Machines	1,406	±6	14,940	—
138	grok-4-fast-reasoning	xAI	1,404	±5	17,965	—
139	gemini-2.5-flash-preview-09-2025	Google	1,404	±4	32,350	—
140	qwen3-235b-a22b-no-thinking	Alibaba	1,402	±5	37,377	—
141	gpt-5.4-nano-high	OpenAI	1,402	±4	58,562	—
142	o1-2024-12-17	OpenAI	1,402	±4	27,807	—
143	Me longcat-flash-chat	meituan	1,402	±6	11,167	—
144	claude-sonnet-4-20250514-thinking-32k	Anthropic	1,401	±4	33,955	—
145	Qwen: Qwen3 235B A22B Thinking 2507	Alibaba	1,400	±7	8,796	—
146	Qwen: Qwen3 Next 80B A3B Instruct	Alibaba	1,399	±5	22,520	—
147	DeepSeek: R1	DeepSeek	1,398	±5	18,524	75
148	Qwen: Qwen3.5-Flash	Alibaba	1,397	±4	58,299	—
149	DeepSeek: DeepSeek V3 0324	DeepSeek	1,396	±4	44,786	64
150	hunyuan-vision-1.5-thinking	Tencent	1,395	±12	2,170	—
151	Qwen: Qwen3.5-35B-A3B	Alibaba	1,395	±4	29,067	—
152	Qwen: Qwen3 VL 235B A22B Thinking	Alibaba	1,395	±7	7,789	—
153	Az amazon-nova-experimental-chat-12-10	Amazon	1,395	±10	3,633	—
154	St StepFun: Step 3.5 Flash	stepfun	1,394	±4	57,225	—
155	Mi mimo-v2-flash (non-thinking)	Xiaomi	1,392	±4	46,161	—
156	o4-mini-2025-04-16	OpenAI	1,391	±4	44,618	—
157	MiniMax: MiniMax M2.5	MiniMax	1,391	±4	40,895	—
158	claude-sonnet-4-20250514	Anthropic	1,390	±4	38,942	—
159	gpt-5-mini-high	OpenAI	1,389	±5	26,551	—
160	o1-preview	OpenAI	1,388	±5	31,122	—
161	claude-3-7-sonnet-20250219-thinking-32k	Anthropic	1,388	±4	38,088	—
162	hunyuan-t1-20250711	Tencent	1,387	±9	4,575	—
163	Mistral: Mistral Medium 3	Mistral	1,387	±5	32,660	—
164	Qwen3-Coder-480B-A35B-Instruct	Alibaba	1,387	±5	25,081	—
165	Mi mimo-v2-flash (thinking)	Xiaomi	1,386	±6	10,751	—
166	minimax-m2.1-preview	MiniMax	1,384	±5	16,638	—
167	hunyuan-turbos-20250416	Tencent	1,383	±7	10,583	—
168	gpt-4.1-mini-2025-04-14	OpenAI	1,382	±4	38,630	—
169	Qwen: Qwen3 30B A3B Instruct 2507	Alibaba	1,382	±5	23,222	—
170	gemini-2.5-flash-lite-preview-09-2025-no-thinking	Google	1,380	±3	46,384	—
171	TL trinity-large-preview	Unknown	1,379	±4	29,770	—
172	Z.ai: GLM 4.6V	Z.ai	1,378	±11	2,760	—
173	solar-pro4	Upstage	1,377	±12	2,437	—
174	Qwen: Qwen3 235B A22B	Alibaba	1,375	±5	25,865	—
175	gemini-2.5-flash-lite-preview-06-17-thinking	Google	1,375	±5	32,209	—
176	qwen2.5-max	Alibaba	1,374	±4	32,417	—
177	claude-3-5-sonnet-20241022	Anthropic	1,374	±3	87,693	45
178	Z.ai: GLM 4.5 Air	Z.ai	1,373	±4	30,368	—
179	claude-3-7-sonnet-20250219	Anthropic	1,372	±4	42,452	74
180	AA Arcee AI: Trinity Large Thinking	arcee-ai	1,369	±5	28,999	—
181	Qwen: Qwen3 Next 80B A3B Thinking	Alibaba	1,369	±6	13,389	—
182	Z.ai: GLM 4.7 Flash	Z.ai	1,367	±6	11,513	—
183	Az amazon-nova-experimental-chat-11-10	Amazon	1,365	±4	25,099	—
184	Google: Gemma 3 27B	Google	1,365	±4	46,434	47
185	MiniMax: MiniMax M1	MiniMax	1,364	±4	34,028	—
186	grok-3-mini-high	xAI	1,364	±5	15,535	—
187	OpenAI: o3 Mini High	OpenAI	1,363	±5	18,589	—
188	nvidia-nemotron-3-super-120b-a12b	NVIDIA	1,361	±7	7,461	—
189	GD gemini-2.0-flash-001	Google DeepMind,Google	1,360	±4	43,346	61
190	DeepSeek: DeepSeek V3	DeepSeek	1,358	±5	21,770	62
191	grok-3-mini-beta	xAI	1,358	±5	20,838	—
192	mistral-small-2506	Mistral	1,357	±5	17,401	—
193	I3 intellect-3	Unknown	1,356	±8	5,282	—
194	Cohere: Command A	Cohere	1,354	±3	55,434	—
195	gemini-2.0-flash-lite-preview-02-05	Google	1,353	±4	24,955	—
196	Z.ai: GLM 4.5V	Z.ai	1,353	±8	4,818	—
197	OpenAI: gpt-oss-120b	OpenAI	1,352	±4	29,935	62
198	nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4	NVIDIA	1,352	±10	4,533	—
199	gemini-1.5-pro-002	Google	1,351	±3	55,606	—
200	Az amazon-nova-experimental-chat-10-20	Amazon	1,350	±6	11,362	—
201	St step-3	stepfun	1,350	±8	6,280	—
202	hunyuan-turbos-20250226	Tencent	1,349	±12	2,220	—
203	OpenAI: o3 Mini	OpenAI	1,348	±4	56,653	86
204	llama-3.1-nemotron-ultra-253b-v1	NVIDIA	1,347	±12	2,549	—
205	Qwen: Qwen3 32B	Alibaba	1,347	±9	3,926	—
206	Az amazon-nova-experimental-chat-10-09	Amazon	1,347	±11	2,776	—
207	In Inception: Mercury 2	inception	1,346	±11	3,043	—
208	qwen-plus-0125	Alibaba	1,346	±8	5,819	—
209	OpenAI: GPT-4o (2024-05-13)	OpenAI	1,346	±3	112,881	—
210	MiniMax: MiniMax M2	MiniMax	1,346	±8	6,748	—
211	AG ling-flash-2.0	ant-group	1,344	±7	6,852	—
212	nvidia-llama-3.3-nemotron-super-49b-v1.5	NVIDIA	1,343	±10	3,316	—
213	claude-3-5-sonnet-20240620	Anthropic	1,343	±3	82,419	—
214	glm-4-plus-0111	Z.ai	1,343	±8	5,760	—
215	Google: Gemma 3 12B	Google	1,342	±10	3,829	—
216	granite-4.2-30b	IBM	1,341	±11	3,287	—
217	hunyuan-turbo-0110	Tencent	1,341	±12	2,290	—
218	gpt-5-nano-high	OpenAI	1,337	±7	8,128	—
219	o1-mini	OpenAI	1,337	±4	51,981	65
220	Az Amazon: Nova 2 Lite	Amazon	1,336	±6	12,117	—
221	QwQ-32B	Alibaba	1,336	±4	25,094	69
222	gemini-advanced-0514	Google	1,336	±5	50,148	—
223	grok-2-2024-08-13	xAI	1,335	±4	63,498	—
224	OpenAI: GPT-4o (2024-08-06)	OpenAI	1,335	±4	45,499	—
225	llama-3.1-405b-instruct-bf16	Meta	1,335	±4	41,375	—
226	St step-2-16k-exp-202412	stepfun	1,333	±9	4,833	—
227	llama-3.1-405b-instruct-fp8	Meta	1,333	±4	59,656	—
228	A2 olmo-3.1-32b-instruct	Allen Institute for AI	1,330	±6	11,463	—
229	YL yi-lightning	Unknown	1,328	±5	27,332	—
230	llama-3.3-nemotron-49b-super-v1	NVIDIA	1,328	±12	2,218	—
231	A2 molmo-2-8b	Allen Institute for AI	1,328	±21	792	—
232	Qwen: Qwen3 30B A3B	Alibaba	1,327	±5	26,093	—
233	llama-4-maverick-17b-128e-instruct	Meta	1,327	±4	39,349	—
234	hunyuan-large-2025-02-10	Tencent	1,326	±10	3,738	—
235	gpt-4-turbo-2024-04-09	OpenAI	1,324	±4	98,114	—
236	claude-3-5-haiku-20241022	Anthropic	1,324	±3	68,972	30
237	gemini-1.5-pro-001	Google	1,324	±4	79,138	—
238	deepseek-v2.5-1210	DeepSeek	1,323	±8	6,795	—
239	gpt-4.1-nano-2025-04-14	OpenAI	1,322	±8	6,103	—
240	claude-3-opus-20240229	Anthropic	1,322	±3	194,909	32
241	Llama-4-Scout-17B-16E-Instruct	Meta	1,321	±5	29,739	—
242	AG ring-flash-2.0	ant-group	1,321	±7	6,904	—
243	St step-1o-turbo-202506	stepfun	1,320	±7	8,856	—
244	glm-4-plus	Z.ai	1,319	±5	26,126	—
245	qwen-max-0919	Alibaba	1,318	±6	16,478	—
246	OpenAI: GPT-4o-mini (2024-07-18)	OpenAI	1,317	±4	68,698	—
247	Meta: Llama 3.3 70B Instruct	Meta	1,317	±4	54,411	34
248	Google: Gemma 3n 4B	Google	1,317	±5	22,084	—
249	OpenAI: gpt-oss-20b	OpenAI	1,317	±6	10,386	42
250	qwen2.5-plus-1127	Alibaba	1,315	±6	10,187	—
251	nvidia-nemotron-3-nano-30b-a3b-bf16	NVIDIA	1,314	±6	15,338	—
252	Mistral Large 2407	Mistral	1,314	±4	45,459	—
253	AV athene-v2-chat	Unknown	1,314	±5	24,739	—
254	gpt-4-0125-preview	OpenAI	1,313	±4	93,439	—
255	gpt-4-1106-preview	OpenAI	1,312	±4	100,105	—
256	hunyuan-standard-2025-02-10	Tencent	1,311	±10	3,904	—
257	gemini-1.5-flash-002	Google	1,309	±4	34,902	—
258	grok-2-mini-2024-08-13	xAI	1,308	±4	52,567	—
259	IA mercury	inception-ai	1,308	±14	1,929	—
260	deepseek-v2.5	DeepSeek	1,307	±5	24,572	—
261	A2 AllenAI: Olmo 3 32B Think	Allen Institute for AI	1,307	±8	5,682	—
262	A7 athene-70b-0725	Unknown	1,306	±6	19,621	—
263	IBM: Granite 4.1 8B	IBM	1,306	±10	4,027	—
264	mistral-large-2411	Mistral	1,305	±4	28,073	37
265	magistral-medium-2506	Mistral	1,304	±6	11,472	—
266	Google: Gemma 3 4B	Google	1,303	±9	4,171	—
267	mistral-small-3.1-24b-instruct-2503	Mistral	1,303	±5	32,621	—
268	Qwen2.5 72B Instruct	Alibaba	1,303	±4	39,406	—
269	llama-3.1-nemotron-70b-instruct	NVIDIA	1,299	±8	7,140	—
270	granite-4.2-3b	IBM	1,295	±12	3,097	—
271	hunyuan-large-vision	Tencent	1,294	±9	5,260	—
272	Meta: Llama 3.1 70B Instruct	Meta	1,293	±4	55,240	—
273	Az amazon.nova-pro-v1:0	Amazon	1,290	±5	24,745	—
274	J1 jamba-1.5-large	Unknown	1,289	±7	8,662	—
275	Google: Gemma 2 27B	Google	1,289	±3	75,754	21
276	IBM: Granite 4.2 8B	IBM	1,288	±11	3,119	—
277	Re reka-core-20240904	Reka AI	1,288	±7	7,312	—
278	gpt-4-0314	OpenAI	1,287	±5	54,173	—
279	gemini-1.5-flash-001	Google	1,286	±5	62,833	—
280	llama-3.1-nemotron-51b-instruct	NVIDIA	1,286	±10	3,749	—
281	A2 llama-3.1-tulu-3-70b	Allen Institute for AI	1,286	±10	2,846	—
282	A2 olmo-3.1-32b-think	Allen Institute for AI	1,286	±7	8,112	—
283	ibm-granite-h-small	IBM	1,285	±8	5,534	—
284	claude-3-sonnet-20240229	Anthropic	1,281	±4	109,284	—
285	gemma-2-9b-it-simpo	Google	1,280	±7	10,072	—
286	nemotron-4-340b-instruct	NVIDIA	1,277	±5	19,659	—
287	llama-3-70b-instruct	Meta	1,276	±4	156,876	—
288	Cohere: Command R+ (08-2024)	Cohere	1,276	±7	9,866	—
289	gpt-4-0613	OpenAI	1,276	±4	88,723	—
290	Mistral: Mistral Small 3	Mistral	1,274	±6	14,681	—
291	glm-4-0520	Z.ai	1,273	±7	9,788	—
292	Re reka-flash-20240904	Reka AI	1,272	±7	7,536	—
293	Qwen2.5 Coder 32B Instruct	Alibaba	1,270	±8	5,432	—
294	c4ai-aya-expanse-32b	Cohere	1,267	±5	27,124	—
295	gemma-2-9b-it	Google	1,267	±4	54,611	13
296	deepseek-coder-v2	DeepSeek	1,265	±6	15,147	—
297	qwen2-72b-instruct	Alibaba	1,261	±5	37,325	—
298	command-r-plus	Cohere	1,261	±4	77,554	—
299	Anthropic: Claude 3 Haiku	Anthropic	1,261	±4	117,701	—
300	Az amazon.nova-lite-v1:0	Amazon	1,260	±5	19,372	—
301	gemini-1.5-flash-8b-001	Google	1,258	±4	35,558	—
302	Microsoft: Phi 4	Microsoft	1,256	±5	24,126	41
303	A2 olmo-2-0325-32b-instruct	Allen Institute for AI	1,251	±11	3,334	—
305	mistral-large-2402	Mistral	1,242	±5	62,436	—
306	Az amazon.nova-micro-v1:0	Amazon	1,240	±5	19,364	—
307	J1 jamba-1.5-mini	Unknown	1,239	±7	8,858	—
308	ministral-8b-2410	Mistral	1,237	±9	4,781	—
309	gemini-pro-dev-api	Google	1,236	±7	18,354	—
310	qwen1.5-110b-chat	Alibaba	1,234	±6	26,195	—
311	hunyuan-standard-256k	Tencent	1,233	±12	2,728	—
312	Re reka-flash-21b-20240226-online	Reka AI	1,233	±7	15,450	—
313	qwen1.5-72b-chat	Alibaba	1,233	±5	39,302	—
314	mixtral-8x22b-instruct-v0.1	Mistral	1,229	±5	51,416	—
315	Cohere: Command R (08-2024)	Cohere	1,226	±5	54,036	—
316	Re reka-flash-21b-20240226	Reka AI	1,226	±6	24,806	—
317	gpt-3.5-turbo-0125	OpenAI	1,225	±5	66,207	—
318	llama-3-8b-instruct	Meta	1,223	±4	104,642	—
319	gemini-pro	Google	1,223	±12	6,390	—
320	c4ai-aya-expanse-8b	Cohere	1,223	±7	9,818	—
321	mistral-medium	Mistral	1,222	±5	34,550	—
322	A2 llama-3.1-tulu-3-8b	Allen Institute for AI	1,220	±11	2,896	—
323	0A Yi-1.5-34B-Chat	01.AI	1,212	±5	24,146	—
324	ZO zephyr-orpo-141b-A35b-v0.1	Unknown	1,212	±11	4,652	—
325	Meta: Llama 3.1 8B Instruct	Meta	1,211	±4	49,605	—
326	granite-3.1-8b-instruct	IBM	1,208	±11	3,090	—
327	gpt-3.5-turbo-1106	OpenAI	1,203	±9	16,619	—
328	qwen1.5-32b-chat	Alibaba	1,203	±6	21,741	—
329	gemma-2-2b-it	Google	1,200	±4	46,616	—
330	phi-3-medium-4k-instruct	Microsoft	1,197	±5	25,055	—
331	Mixtral-8x7B-Instruct-v0.1	Mistral	1,197	±4	73,503	—
332	DI dbrx-instruct-preview	Unknown	1,195	±6	32,191	—
333	qwen1.5-14b-chat	Alibaba	1,191	±7	17,839	—
334	I5 internlm2_5-20b-chat	Unknown	1,190	±7	9,901	—
335	deepseek-llm-67b-chat	DeepSeek	1,184	±11	4,932	—
336	wizardlm-70b	Microsoft	1,184	±9	8,214	—
337	0A Yi-34B-Chat	01.AI	1,183	±7	15,483	—
338	granite-3.0-8b-instruct	IBM	1,182	±9	6,638	—
339	O3 openchat-3.5	Unknown	1,182	±10	7,968	—
340	O3 openchat-3.5-0106	Unknown	1,182	±8	12,637	—
341	gemma-1.1-7b-it	Google	1,182	±6	23,893	—
342	snowflake-arctic-instruct	Snowflake	1,179	±6	32,832	—
343	granite-3.1-2b-instruct	IBM	1,178	±11	3,188	—
344	T2 tulu-2-dpo-70b	Unknown	1,177	±10	6,535	—
345	O2 openhermes-2.5-mistral-7b	Unknown	1,175	±10	5,006	—
346	V3 vicuna-33b	Unknown	1,172	±6	22,479	—
347	Phi-3-small-8k-instruct	Microsoft	1,171	±6	17,766	—
348	SL starling-lm-7b-beta	Unknown	1,171	±7	16,056	—
349	llama-2-70b-chat	Meta	1,170	±5	38,492	—
350	SL starling-lm-7b-alpha	Unknown	1,167	±8	10,224	—
351	Meta: Llama 3.2 3B Instruct	Meta	1,166	±8	7,936	—
352	nous-hermes-2-mixtral-8x7b-dpo	Nous Research	1,164	±12	3,777	—
353	granite-3.0-2b-instruct	IBM	1,156	±8	6,837	—
354	llama2-70b-steerlm-chat	NVIDIA	1,154	±13	3,585	—
355	QwQ-32B-Preview	Alibaba	1,154	±11	3,231	—
356	solar-10.7b-instruct-v1.0	Upstage	1,152	±13	4,155	—
357	D2 dolphin-2.2.1-mistral-7b	Unknown	1,152	±15	1,679	—
358	M3 mpt-30b-chat	Unknown	1,150	±12	2,572	—
359	wizardlm-13b	Microsoft	1,149	±9	7,044	—
360	mistral-7b-instruct-v0.2	Mistral	1,149	±7	19,402	—
361	F1 falcon-180b-chat	Unknown	1,147	±17	1,295	—
362	qwen1.5-7b-chat	Alibaba	1,143	±10	4,737	—
363	phi-3-mini-4k-instruct-june-2024	Microsoft	1,143	±6	12,297	—
364	llama-2-13b-chat	Meta	1,141	±7	19,174	—
365	V1 vicuna-13b	Unknown	1,141	±7	19,367	—
366	qwen-14b-chat	Alibaba	1,139	±11	4,964	—
367	palm-2	Google	1,138	±9	8,554	—
368	gemma-7b-it	Google	1,137	±9	8,925	—
369	codellama-34b-instruct	Meta	1,136	±9	7,366	—
370	Z7 zephyr-7b-beta	Unknown	1,130	±9	11,118	—
371	phi-3-mini-128k-instruct	Microsoft	1,129	±7	20,685	—
372	Phi-3-mini-4k-instruct	Microsoft	1,128	±6	20,118	—
373	G3 guanaco-33b	Unknown	1,127	±12	2,921	—
374	Z7 zephyr-7b-alpha	Unknown	1,126	±16	1,785	—
375	SN stripedhyena-nous-7b	Unknown	1,121	±11	5,182	—
376	codellama-70b-instruct	Meta	1,119	±18	1,143	—
377	gemma-1.1-2b-it	Google	1,116	±8	10,854	—
378	V7 vicuna-7b	Unknown	1,115	±9	6,923	—
379	S1 smollm2-1.7b-instruct	Unknown	1,114	±14	2,199	—
380	Meta: Llama 3.2 1B Instruct	Meta	1,111	±8	8,045	—
381	mistral-7b-instruct	Mistral	1,110	±9	8,977	—
382	llama-2-7b-chat	Meta	1,107	±7	14,148	—
383	gemma-2b-it	Google	1,093	±11	4,780	—
384	qwen1.5-4b-chat	Alibaba	1,090	±9	7,597	—
385	A2 olmo-7b-instruct	Allen Institute for AI	1,073	±11	6,328	—
386	K1 koala-13b	Unknown	1,070	±10	6,965	—
387	A1 alpaca-13b	Unknown	1,069	±11	5,745	—
388	G1 gpt4all-13b-snoozy	Unknown	1,067	±15	1,743	—
389	M7 mpt-7b-chat	Unknown	1,063	±12	3,924	—
390	C6 chatglm3-6b	Unknown	1,056	±12	4,658	—
391	R4 RWKV-4-Raven-14B	Unknown	1,042	±11	4,845	—
392	C6 chatglm2-6b	Unknown	1,024	±14	2,658	—
393	OP oasst-pythia-12b	Unknown	1,023	±11	6,310	—
394	chatglm-6b	Z.ai	995	±13	4,914	—
395	FT fastchat-t5-3b	Unknown	992	±12	4,203	—
396	DV dolly-v2-12b	Unknown	982	±13	3,412	—
397	llama-13b	Meta	974	±16	2,391	—
398	ST stablelm-tuned-alpha-7b	Unknown	953	±13	3,287	—

The Known Good

How to read Elo. A rating is only meaningful against the population that produced it: a model's number in one arena cannot be compared to its number in another, and a rating built on a few hundred votes moves far more than one built on tens of thousands — the 95% confidence interval is the honest guide to that. Elo measures which output a person preferred, not whether it was correct; for correctness see the evaluations and the Known Good Index, which does not include Elo. Ratings are ingested from Arena and are published under the source's own terms — see attribution.