Ran it, or it didn’t happen.

Benchmark open models on your own hardware. Runs count once their event log checks out.

Recent verified runsWhereSpeed

PassedFailedNot supported

Cloud leaderboard

Every model runs on the same pinned machine. 17 models, 70 runs.

Suite: [email protected]
Machine: NVIDIA L40S · 48 GB
Engine: ollama 0.40.14

#	Model	Pass rate				Speed	Runs
1	Qwen3.5-35B-A3BQ4_K_M	96%	9/9	8/9	9/9	201.4 tok/s	3
2	Qwen3.5-27BQ4_K_M	93%	8/9	9/9	8/9	35.5 tok/s	3
3	Qwen3-30B-A3B-Instruct-2507Q4_K_M	92%	11/12	10/12	12/12	187.6 tok/s	4
4	GLM-4.7-FlashQ4_K_M	89%	9/9	6/9	9/9	201.7 tok/s	3
5	Qwen3-Coder-30B-A3B-InstructQ4_K_M	89%	9/9	7/9	8/9	186.4 tok/s	3
6	Qwen3-8BQ4_K_M	89%	9/12	11/12	12/12	120.7 tok/s	4
7	Qwen3-14BQ4_K_M	87%	13/15	11/15	15/15	67.1 tok/s	5
8	gemma-3-27b-itQ4_K_M	86%	17/18	Not supported	14/18	36.2 tok/s	6
9	gpt-oss-20bMXFP4	85%	9/9	9/9	5/9	173.8 tok/s	3
10	gemma-3-4b-itQ4_K_M	83%	11/15	Not supported	14/15	232.7 tok/s	5
11	Qwen3.5-2BQ4_K_M	81%	9/12	9/12	11/12	298.1 tok/s	4
12	Qwen3-4BQ4_K_M	78%	6/9	6/9	9/9	247.9 tok/s	3
13	gemma-3-12b-itQ4_K_M	75%	8/12	Not supported	10/12	80.9 tok/s	4
14	Llama-3.1-8B-InstructQ4_K_M	71%	8/15	9/15	15/15	124.4 tok/s	5
15	Phi-3.5-mini-instructQ4_K_M	64%	9/18	Not supported	14/18	261.3 tok/s	6
16	MiniCPM5-2BQ4_K_M	52%	4/9	3/9	7/9	301.4 tok/s	3
17	Mistral-7B-Instruct-v0.2Q4_K_M	47%	6/18	Not supported	11/18	137.1 tok/s	6

Ranked by checks passed. A task a model can’t do is shown, not counted as a zero.

Open model leaderboard

144 open-weight models in non-english ranked by LMArena score · updated Oct 28, 2026

Top rated

Most downloaded

Newest

All sizes ≤3B 3–9B 9–20B 20–40B 40–80B >80B

#	Model	Score	Votes	Params	License	Downloads	Likes
1	Kimi-K3moonshotai	1466	22.9K	2.8T	Custom	771.9K	11,938
2	GLM-5.3-Flashzai-org	1463	26.4K	321B	MIT	6.3M	3,066
3	GLM-5.3zai-org	1460	15.9K	753B	Custom	2.2M	2,384
4	GLM-5.2zai-org	1458	31.8K	753B	MIT	324.4K	5,202
5	GLM-5.1zai-org	1449	38.4K	754B	MIT	918.3K	1,850
6	MiMo-V2.5-ProXiaomiMiMo	1448	45.3K	1.0T	MIT	9.9K	771
7	Kimi-K2.6moonshotai	1446	21.8K	1.0T	Modified MIT	238.6K	1,639
8	DeepSeek-V4-Prodeepseek-ai	1439	32.3K	1.6T	MIT	333.6K	5,700
9	Inklingthinkingmachines	1435	26.9K	952B	Apache 2.0	1.8M	1,837
10	Kimi-K2.5moonshotai	1432	42.9K	1.0T	Modified MIT	394.2K	2,898
11	gemma-4-31Bgoogle	1431	3.8K	31.3B	Apache 2.0	61K	589
12	GLM-5zai-org	1430	16.3K	754B	MIT	1.5M	2,141
13	Qwen3.5-397B-A17BQwen	1430	54.9K	403B	Apache 2.0	881.4K	1,581
14	Hy3tencent	1428	8.6K	299B	Apache 2.0	1.2M	990
15	NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16nvidia	1427	6.2K	561B	OpenMDW 1.1	1.2M	382
16	Qwen3.8-27BQwen	1426	17.4K	27.8B	Apache 2.0	7.8M	19,243
17	GLM-4.6zai-org	1426	20.5K	357B	MIT	26.9K	1,247
18	gemma-4-26B-A4Bgoogle	1422	3.8K	25.8B	Apache 2.0	72.3K	463
19	DeepSeek-V4-Flashdeepseek-ai	1419	28.8K	291B	MIT	1.1M	2,378
20	MiniMax-M3MiniMaxAI	1418	40.5K	427B	Custom	158.6K	1,601
21	GLM-4.7zai-org	1417	7.2K	358B	MIT	68.4K	2,077
22	GLM-4.5zai-org	1417	13.7K	358B	MIT	109.4K	1,439
23	Mistral-Large-3-675B-Instruct-2512mistralai	1414	51.8K	—	Apache 2.0	2.5K	261
24	Beam-1.0-501B-A23Breflection	1412	2.6K	501B	Apache 2.0	11.4K	305
25	DeepSeek-R1-0528deepseek-ai	1412	10K	685B	MIT	121K	2,494
26	Qwen3-235B-A22B-Instruct-2507Qwen	1409	57K	235B	Apache 2.0	86.1K	814
27	DeepSeek-V3.2deepseek-ai	1409	28.5K	685B	MIT	7.3M	1,594
28	DeepSeek-V3.1-Terminusdeepseek-ai	1407	2K	685B	MIT	9.8K	386
29	Qwen3-Next-80B-A3B-InstructQwen	1407	13K	81.3B	Apache 2.0	129.4K	1,072
30	MiMo-V2.5XiaomiMiMo	1405	26K	311B	MIT	111.4K	434
31	Mistral-Medium-3.5-128Bmistralai	1405	6.1K	128B	Custom	19.4K	497
32	Qwen3-VL-235B-A22B-InstructQwen	1404	6.4K	236B	Apache 2.0	323.5K	430
33	DeepSeek-V3.2-Expdeepseek-ai	1404	5.1K	685B	MIT	42.7K	1,027
34	LongCat-Flash-Chatmeituan-longcat	1404	6.4K	562B	MIT	13.5K	540
35	Inkling-Smallthinkingmachines	1403	22K	266B	Apache 2.0	1.3M	443
36	Qwen3.5-122B-A10BQwen	1401	16.6K	125B	Apache 2.0	942.5K	653
37	DeepSeek-V3.1deepseek-ai	1400	6.8K	685B	MIT	260K	864
38	Qwen3-235B-A22B-Thinking-2507Qwen	1397	4.9K	235B	Apache 2.0	9.1K	419
39	MiMo-V2-FlashXiaomiMiMo	1391	28.1K	310B	MIT	22.8K	756
40	Qwen3.5-27BQwen	1390	15.9K	27.8B	Apache 2.0	1.7M	1,088
41	Hy3-previewtencent	1390	3.5K	299B	Custom	67K	314
42	Step-3.5-Flashstepfun-ai	1385	35.1K	199B	Apache 2.0	220.3K	847
43	Muse-Glimmer-30Bmeta-models	1385	2.3K	29.8B	Apache 2.0	185K	2,124
44	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16nvidia	1384	2.6K	124B	OpenMDW 1.1	68.3K	115
45	Qwen3-235B-A22BQwen	1384	20.9K	235B	Apache 2.0	295.6K	1,128
46	MiniMax-M2.7MiniMaxAI	1383	52.6K	229B	Custom	314.4K	1,261
47	Qwen3.5-35B-A3BQwen	1378	16.9K	36.0B	Apache 2.0	1.7M	1,554
48	Qwen3-VL-235B-A22B-ThinkingQwen	1377	4.4K	236B	Apache 2.0	80.8K	420
49	Qwen3-30B-A3B-Instruct-2507Qwen	1372	13.5K	30.5B	Apache 2.0	935K	855
50	GLM-4.5-Airzai-org	1366	17.4K	110B	MIT	1.3M	649
51	GLM-4.6Vzai-org	1359	1.7K	108B	MIT	5.4K	398
52	DeepSeek-V3-0324deepseek-ai	1358	23.6K	685B	MIT	157.7K	3,223
53	DeepSeek-R1deepseek-ai	1356	7.8K	684B	MIT	1.6M	14,434
54	NVIDIA-Nemotron-3-Super-120B-A12B-BF16nvidia	1356	4.7K	124B	Custom	1.1M	438
55	gpt-oss-120bopenai	1350	17.1K	117B	Apache 2.0	2.4M	5,496
56	Qwen3-Coder-480B-A35B-InstructQwen	1345	14.2K	480B	Apache 2.0	57.8K	1,385
57	Qwen3-Next-80B-A3B-ThinkingQwen	1342	7.6K	81.3B	Apache 2.0	43.8K	503
58	granite-4.2-30bibm-granite	1341	1.9K	29.3B	Apache 2.0	30.4K	140
59	MiniMax-M2.5MiniMaxAI	1338	24K	229B	Modified MIT	1.3M	1,529
60	Ling-flash-2.0inclusionAI	1335	3.9K	103B	MIT	4.3K	222
61	gemma-3-27b-itgoogle	1334	23.6K	27.4B	Gemma	361.1K	2,066
62	GLM-4.7-Flashzai-org	1330	7.4K	31.2B	MIT	1M	1,895
63	INTELLECT-3PrimeIntellect	1328	3K	107B	MIT	768	218
64	Trinity-Large-Thinkingarcee-ai	1325	15.9K	399B	OpenMDW 1.1	3.1K	190
65	NVIDIA-Nemotron-3-Nano-30B-A3B-BF16nvidia	1320	9.3K	31.6B	Custom	1.6M	840
66	Trinity-Large-Previewarcee-ai	1318	17.3K	399B	OpenMDW 1.1	1.6K	181
67	gemma-3-12b-itgoogle	1318	1.5K	12.2B	Gemma	360.2K	907
68	Qwen3-32BQwen	1317	1.7K	32.8B	Apache 2.0	2.9M	791
69	DeepSeek-V3deepseek-ai	1315	8.9K	684B	—	1.6M	4,462
70	Mistral-Small-3.2-24B-Instruct-2506mistralai	1315	9.9K	24.0B	Apache 2.0	110.3K	644
71	MiniMax-M2MiniMaxAI	1313	3.8K	229B	Modified MIT	106.3K	1,514
72	QwQ-32BQwen	1305	12.2K	32.8B	Apache 2.0	32.1K	2,970
73	GLM-4.5Vzai-org	1303	2.8K	108B	MIT	16.8K	726
74	granite-4.2-8bibm-granite	1300	1.8K	8.8B	Apache 2.0	68.3K	115
75	Ring-flash-2.0inclusionAI	1299	3.9K	103B	MIT	1.3K	104
76	NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16nvidia	1298	14.6K	31.6B	OpenMDW 1.1	1.1M	269
77	Qwen3-30B-A3BQwen	1295	13.8K	30.5B	Apache 2.0	1.1M	982
78	gemma-3n-E4B-itgoogle	1285	12K	7.8B	Gemma	12.6K	938
79	Olmo-3.1-32B-Instructallenai	1275	7.2K	32.2B	Apache 2.0	20.4K	88
80	Molmo2-8Ballenai	1274	492	8.7B	Apache 2.0	43.5K	198
81	DeepSeek-V2.5-1210deepseek-ai	1273	3K	236B	Custom	2K	275
82	gemma-3-4b-itgoogle	1273	1.6K	4.3B	Gemma	996.3K	1,629
83	Llama-4-Maverick-17B-128E-Instructmeta-llama	1269	20.5K	402B	Llama 4	11.6K	520
84	gpt-oss-20bopenai	1268	5.9K	20.9B	Apache 2.0	5.9M	5,275
85	granite-4.2-3bibm-granite	1268	1.9K	3.7B	Apache 2.0	37.3K	129
86	granite-4.1-8bibm-granite	1260	2.2K	8.8B	Apache 2.0	98.3K	257
87	Mistral-Small-3.1-24B-Instruct-2503mistralai	1254	18.5K	24.0B	Apache 2.0	169.9K	1,408
88	Olmo-3-32B-Thinkallenai	1254	3.4K	32.2B	Apache 2.0	12.1K	177
89	DeepSeek-V2.5deepseek-ai	1253	11.7K	236B	Custom	4.5K	750
90	Llama-4-Scout-17B-16E-Instructmeta-llama	1252	16.8K	109B	Llama 4	109.2K	1,386
91	Qwen2.5-72B-InstructQwen	1252	18.8K	72.7B	Custom	185.7K	1,042
92	Llama-3.1-405B-Instructmeta-llama	1248	27.2K	406B	Llama 3.1	9.2K	607
93	Llama-3.1-Nemotron-70B-Instructnvidia	1245	3.4K	—	Llama 3.1	72	570
94	Llama-3.3-70B-Instructmeta-llama	1236	25.2K	70.6B	Llama 3.3	357.9K	3,193
95	Llama-3.1-Tulu-3-70Ballenai	1235	1.2K	70.6B	Llama 3.1	947	61
96	Llama-3.1-70B-Instructmeta-llama	1219	25.5K	70.6B	Llama 3.1	91.6K	1,027
97	gemma-2-27b-itgoogle	1217	34.5K	27.2B	Gemma	20.5K	577
98	granite-4.0-h-smallibm-granite	1216	3.1K	32.2B	Apache 2.0	23.3K	313
99	aya-expanse-32bCohereLabs	1213	12.8K	32.3B	CC BY-NC 4.0	3K	302
100	Olmo-3.1-32B-Thinkallenai	1209	5K	32.2B	Apache 2.0	6.7K	119
101	Nemotron-4-340B-Instructnvidia	1206	9.4K	—	Custom	654	703
102	Qwen2.5-Coder-32B-InstructQwen	1204	2.8K	32.8B	Apache 2.0	546K	2,205
103	Mistral-Small-24B-Instruct-2501mistralai	1198	5.9K	23.6B	Apache 2.0	94K	991
104	phi-4microsoft	1197	10K	14.7B	MIT	299.1K	2,338
105	gemma-2-9b-itgoogle	1188	24.8K	9.2B	Gemma	119K	1,165
106	DeepSeek-Coder-V2-Instructdeepseek-ai	1182	7K	236B	Custom	3.8K	718
107	aya-expanse-8bCohereLabs	1180	4.3K	8.0B	CC BY-NC 4.0	24.4K	463
108	Qwen2-72B-InstructQwen	1176	17.4K	72.7B	Tongyi Qianwen	24K	724
109	Llama-3.1-Tulu-3-8Ballenai	1169	1.3K	8.0B	Llama 3.1	5K	178
110	Llama-3.1-8B-Instructmeta-llama	1148	23K	8.0B	Llama 3.1	4.3M	8,816
111	Qwen1.5-110B-ChatQwen	1142	12.3K	111B	Tongyi Qianwen	2.7K	139
112	Meta-Llama-3-70B-Instructmeta-llama	1142	66.3K	70.6B	Llama 3	37.1K	1,537
113	Qwen1.5-72B-ChatQwen	1135	14.3K	72.3B	Tongyi Qianwen	11.1K	224
114	QwQ-32B-PreviewQwen	1129	1.4K	32.8B	Apache 2.0	7.2K	1,747
115	Mixtral-8x22B-Instruct-v0.1mistralai	1128	21.5K	141B	Apache 2.0	38.4K	780
116	gemma-2-2b-itgoogle	1121	20.9K	2.6B	Gemma	129.8K	1,605
117	Qwen1.5-32B-ChatQwen	1106	9.9K	32.5B	Tongyi Qianwen	46.7K	115
118	granite-3.1-8b-instructibm-granite	1099	1.4K	8.2B	Apache 2.0	27.9K	169
119	Meta-Llama-3-8B-Instructmeta-llama	1097	42.9K	8.0B	Llama 3	354.8K	5,756
120	Qwen1.5-14B-ChatQwen	1095	8.3K	14.2B	Tongyi Qianwen	12.1K	118
121	Phi-3-medium-4k-instructmicrosoft	1094	11.8K	14.0B	MIT	18.2K	231
122	Mixtral-8x7B-Instruct-v0.1mistralai	1077	27.2K	46.7B	Apache 2.0	292.7K	4,777
123	deepseek-llm-67b-chatdeepseek-ai	1073	1.1K	—	Custom	2.7K	222
124	granite-3.1-2b-instructibm-granite	1068	1.4K	2.5B	Apache 2.0	24.4K	57
125	Qwen1.5-7B-ChatQwen	1058	1.5K	7.7B	Tongyi Qianwen	7.1K	192
126	Phi-3-small-8k-instructmicrosoft	1058	8.7K	7.4B	MIT	7.1K	182
127	gemma-1.1-7b-itgoogle	1051	11.2K	8.5B	Gemma	4.3K	276
128	Llama-2-70b-chat-hfmeta-llama	1045	12.4K	69.0B	Llama 2	4.6K	2,211
129	Qwen-14B-ChatQwen	1041	1.1K	14.2B	—	5.2K	381
130	granite-3.0-8b-instructibm-granite	1037	3.3K	8.2B	Apache 2.0	13.1K	209
131	granite-3.0-2b-instructibm-granite	1032	3.5K	2.6B	Apache 2.0	3.2K	48
132	Llama-2-13b-chat-hfmeta-llama	1024	5.3K	13.0B	Llama 2	152.7K	1,138
133	Phi-3-mini-4k-instructmicrosoft	1021	9.8K	3.8B	MIT	324.8K	1,489
134	Llama-3.2-3B-Instructmeta-llama	1019	3.8K	3.2B	Llama 3.2	929.3K	2,952
135	SOLAR-10.7B-Instruct-v1.0upstage	1014	878	10.7B	CC BY-NC 4.0	48.9K	660
136	Mistral-7B-Instruct-v0.2mistralai	1012	6.7K	7.2B	Apache 2.0	1.6M	3,279
137	Phi-3-mini-128k-instructmicrosoft	1000	8.1K	3.8B	MIT	236.2K	1,713
138	gemma-7b-itgoogle	998	3.2K	8.5B	Gemma	25.3K	1,263
139	gemma-1.1-2b-itgoogle	988	4.9K	2.5B	Gemma	6.1K	175
140	Qwen1.5-4B-ChatQwen	979	2.8K	4.0B	Custom	15.9K	53
141	OLMo-7B-Instructallenai	977	2.3K	6.9B	Apache 2.0	845	54
142	Llama-3.2-1B-Instructmeta-llama	973	3.7K	1.2B	Llama 3.2	1.5M	1,958
143	Llama-2-7b-chat-hfmeta-llama	973	3.9K	6.7B	Llama 2	126.9K	4,949
144	gemma-2b-itgoogle	958	1.8K	2.5B	Gemma	13.8K	975

Show the top 20

Scores: LMArena, CC BY 4.0. Measured on served versions, not local builds. A model with several served variants shows its best.