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

139 open-weight models in chinese ranked by LMArena score · updated Nov 6, 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	GLM-5.3zai-org	1533	2.4K	753B	Custom	2.1M	2,395
2	GLM-5.3-Flashzai-org	1529	3.9K	321B	MIT	6.4M	3,088
3	Kimi-K3moonshotai	1527	3K	2.8T	Custom	771.6K	11,950
4	GLM-5.2zai-org	1524	4K	753B	MIT	318.5K	5,205
5	Kimi-K2.6moonshotai	1519	2.3K	1.0T	Modified MIT	237.2K	1,641
6	GLM-5.1zai-org	1515	4.8K	754B	MIT	892.7K	1,850
7	GLM-5zai-org	1511	2K	754B	MIT	1.4M	2,141
8	MiMo-V2.5-ProXiaomiMiMo	1506	5K	1.0T	MIT	9.9K	771
9	GLM-4.6zai-org	1499	2.7K	357B	MIT	28.1K	1,247
10	Qwen3.5-397B-A17BQwen	1498	7K	403B	Apache 2.0	857.2K	1,581
11	NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16nvidia	1497	725	561B	OpenMDW 1.1	1.1M	384
12	Inklingthinkingmachines	1495	3.9K	952B	Apache 2.0	1.8M	1,842
13	gemma-4-26B-A4Bgoogle	1495	524	25.8B	Apache 2.0	74.1K	465
14	Kimi-K2.5moonshotai	1495	4.6K	1.0T	Modified MIT	395K	2,901
15	GLM-4.7zai-org	1495	1K	358B	MIT	66.6K	2,077
16	Qwen3.8-27BQwen	1492	2.5K	27.8B	Apache 2.0	7.8M	19,348
17	Hy3tencent	1486	1.1K	299B	Apache 2.0	1.2M	990
18	DeepSeek-V4-Prodeepseek-ai	1481	706	1.6T	MIT	335.6K	5,699
19	Qwen3-235B-A22B-Thinking-2507Qwen	1481	536	235B	Apache 2.0	8.9K	419
20	Qwen3.5-27BQwen	1478	2.1K	27.8B	Apache 2.0	1.7M	1,090
21	gemma-4-31Bgoogle	1475	522	31.3B	Apache 2.0	57K	589
22	MiMo-V2.5XiaomiMiMo	1470	2.6K	311B	MIT	110.3K	434
23	DeepSeek-V3.1deepseek-ai	1468	1.1K	685B	MIT	262.8K	864
24	GLM-4.5zai-org	1465	1.8K	358B	MIT	110.4K	1,439
25	Qwen3-235B-A22B-Instruct-2507Qwen	1464	7.3K	235B	Apache 2.0	84.8K	814
26	MiniMax-M3MiniMaxAI	1464	4.8K	427B	Custom	155.9K	1,600
27	DeepSeek-V4-Flashdeepseek-ai	1464	3K	291B	MIT	1.1M	2,378
28	Inkling-Smallthinkingmachines	1463	3.1K	266B	Apache 2.0	1.2M	449
29	MiMo-V2-FlashXiaomiMiMo	1462	3.4K	310B	MIT	22.2K	756
30	Qwen3.5-122B-A10BQwen	1462	2.1K	125B	Apache 2.0	924.6K	653
31	Qwen3-VL-235B-A22B-InstructQwen	1462	714	236B	Apache 2.0	327K	431
32	DeepSeek-V3.2-Expdeepseek-ai	1462	1K	685B	MIT	43.6K	1,027
33	Qwen3-Next-80B-A3B-InstructQwen	1460	1.7K	81.3B	Apache 2.0	129.9K	1,072
34	Qwen3.5-35B-A3BQwen	1457	2.1K	36.0B	Apache 2.0	1.6M	1,555
35	DeepSeek-V3.2deepseek-ai	1455	3.7K	685B	MIT	7.5M	1,596
36	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16nvidia	1452	283	124B	OpenMDW 1.1	70.3K	116
37	LongCat-Flash-Chatmeituan-longcat	1449	865	562B	MIT	13.3K	540
38	Step-3.5-Flashstepfun-ai	1446	3.7K	199B	Apache 2.0	219K	847
39	Mistral-Large-3-675B-Instruct-2512mistralai	1446	6.9K	—	Apache 2.0	2.5K	261
40	Mistral-Medium-3.5-128Bmistralai	1445	611	128B	Custom	17.2K	498
41	DeepSeek-R1-0528deepseek-ai	1442	1.2K	685B	MIT	122.3K	2,495
42	Qwen3-VL-235B-A22B-ThinkingQwen	1441	444	236B	Apache 2.0	81K	420
43	MiniMax-M2.7MiniMaxAI	1440	6K	229B	Custom	307.1K	1,261
44	Hy3-previewtencent	1434	455	299B	Custom	68.3K	314
45	Qwen3-30B-A3B-Instruct-2507Qwen	1433	1.8K	30.5B	Apache 2.0	932.4K	855
46	NVIDIA-Nemotron-3-Super-120B-A12B-BF16nvidia	1432	586	124B	Custom	1.1M	438
47	GLM-4.5-Airzai-org	1426	2.2K	110B	MIT	1.3M	649
48	Qwen3-235B-A22BQwen	1424	2.6K	235B	Apache 2.0	308.9K	1,128
49	GLM-4.6Vzai-org	1424	276	108B	MIT	5.5K	398
50	Beam-1.0-501B-A23Breflection	1423	212	501B	Apache 2.0	11.8K	311
51	granite-4.2-30bibm-granite	1419	234	29.3B	Apache 2.0	32.9K	140
52	Ling-flash-2.0inclusionAI	1418	454	103B	MIT	4.4K	222
53	Muse-Glimmer-30Bmeta-models	1410	342	29.8B	Apache 2.0	187.5K	2,127
54	Qwen3-Next-80B-A3B-ThinkingQwen	1407	1.1K	81.3B	Apache 2.0	44.4K	503
55	GLM-4.7-Flashzai-org	1404	821	31.2B	MIT	1M	1,896
56	DeepSeek-R1deepseek-ai	1400	1.2K	684B	MIT	1.6M	14,439
57	Ring-flash-2.0inclusionAI	1399	442	103B	MIT	1.4K	104
58	MiniMax-M2.5MiniMaxAI	1394	2.9K	229B	Modified MIT	1.2M	1,529
59	DeepSeek-V3-0324deepseek-ai	1390	3K	685B	MIT	153.8K	3,223
60	gpt-oss-120bopenai	1385	2.3K	117B	Apache 2.0	2.4M	5,498
61	NVIDIA-Nemotron-3-Nano-30B-A3B-BF16nvidia	1383	1.3K	31.6B	Custom	1.6M	840
62	QwQ-32BQwen	1378	1.5K	32.8B	Apache 2.0	32K	2,971
63	Trinity-Large-Thinkingarcee-ai	1372	2.1K	399B	OpenMDW 1.1	3.2K	190
64	MiniMax-M2MiniMaxAI	1366	452	229B	Modified MIT	109.4K	1,514
65	NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16nvidia	1365	2K	31.6B	OpenMDW 1.1	1.1M	271
66	granite-4.2-8bibm-granite	1363	230	8.8B	Apache 2.0	69.2K	116
67	Qwen3-32BQwen	1357	187	32.8B	Apache 2.0	3M	791
68	Qwen3-Coder-480B-A35B-InstructQwen	1357	1.8K	480B	Apache 2.0	58.6K	1,387
69	INTELLECT-3PrimeIntellect	1354	519	107B	MIT	774	218
70	Trinity-Large-Previewarcee-ai	1353	2.1K	399B	OpenMDW 1.1	1.7K	181
71	Qwen3-30B-A3BQwen	1347	1.6K	30.5B	Apache 2.0	1.1M	982
72	gemma-3-27b-itgoogle	1346	2.9K	27.4B	Gemma	364.5K	2,066
73	Mistral-Small-3.2-24B-Instruct-2506mistralai	1340	1.1K	24.0B	Apache 2.0	109.2K	644
74	DeepSeek-V3deepseek-ai	1338	1.5K	684B	—	1.6M	4,462
75	GLM-4.5Vzai-org	1337	235	108B	MIT	16.2K	726
76	granite-4.1-8bibm-granite	1334	220	8.8B	Apache 2.0	98.1K	257
77	DeepSeek-V2.5-1210deepseek-ai	1318	446	236B	Custom	2.1K	275
78	gpt-oss-20bopenai	1315	664	20.9B	Apache 2.0	5.9M	5,286
79	gemma-3n-E4B-itgoogle	1309	1.5K	7.8B	Gemma	12.7K	938
80	Olmo-3.1-32B-Instructallenai	1304	879	32.2B	Apache 2.0	20.8K	88
81	Olmo-3-32B-Thinkallenai	1298	487	32.2B	Apache 2.0	12.2K	177
82	DeepSeek-V2.5deepseek-ai	1281	2.2K	236B	Custom	4.5K	750
83	Llama-4-Maverick-17B-128E-Instructmeta-llama	1277	2.6K	402B	Llama 4	11.7K	520
84	granite-4.2-3bibm-granite	1275	206	3.7B	Apache 2.0	38.9K	131
85	Qwen2.5-72B-InstructQwen	1272	3.3K	72.7B	Custom	192.8K	1,043
86	Llama-3.1-Nemotron-70B-Instructnvidia	1263	723	—	Llama 3.1	74	570
87	Llama-4-Scout-17B-16E-Instructmeta-llama	1254	2.1K	109B	Llama 4	110.7K	1,387
88	Mistral-Small-3.1-24B-Instruct-2503mistralai	1252	2.2K	24.0B	Apache 2.0	168.4K	1,408
89	granite-4.0-h-smallibm-granite	1249	344	32.2B	Apache 2.0	23.5K	313
90	Llama-3.1-Tulu-3-70Ballenai	1249	224	70.6B	Llama 3.1	965	61
91	Llama-3.1-405B-Instructmeta-llama	1242	3.1K	406B	Llama 3.1	9.2K	607
92	Olmo-3.1-32B-Thinkallenai	1241	727	32.2B	Apache 2.0	7K	120
93	Qwen2-72B-InstructQwen	1240	4.1K	72.7B	Tongyi Qianwen	24.3K	724
94	Qwen2.5-Coder-32B-InstructQwen	1222	528	32.8B	Apache 2.0	548.8K	2,206
95	gemma-2-27b-itgoogle	1221	6.9K	27.2B	Gemma	21.7K	577
96	QwQ-32B-PreviewQwen	1218	206	32.8B	Apache 2.0	7.1K	1,747
97	Llama-3.3-70B-Instructmeta-llama	1217	3.7K	70.6B	Llama 3.3	357.4K	3,197
98	Llama-3.1-70B-Instructmeta-llama	1215	4.8K	70.6B	Llama 3.1	92K	1,027
99	Nemotron-4-340B-Instructnvidia	1214	2.4K	—	Custom	638	703
100	phi-4microsoft	1212	1.5K	14.7B	MIT	293.7K	2,340
101	aya-expanse-32bCohereLabs	1211	2.4K	32.3B	CC BY-NC 4.0	3K	303
102	Qwen1.5-110B-ChatQwen	1206	2.9K	111B	Tongyi Qianwen	2.7K	139
103	Mistral-Small-24B-Instruct-2501mistralai	1204	1.1K	23.6B	Apache 2.0	94.4K	991
104	DeepSeek-Coder-V2-Instructdeepseek-ai	1201	1.7K	236B	Custom	3.8K	718
105	Qwen1.5-72B-ChatQwen	1186	4.2K	72.3B	Tongyi Qianwen	11.2K	224
106	gemma-2-9b-itgoogle	1185	5K	9.2B	Gemma	113.6K	1,166
107	aya-expanse-8bCohereLabs	1181	618	8.0B	CC BY-NC 4.0	22.9K	463
108	Qwen1.5-32B-ChatQwen	1177	3.2K	32.5B	Tongyi Qianwen	46.8K	115
109	Llama-3.1-Tulu-3-8Ballenai	1176	238	8.0B	Llama 3.1	5K	178
110	Llama-3.1-8B-Instructmeta-llama	1151	4.4K	8.0B	Llama 3.1	4.3M	8,836
111	Qwen1.5-14B-ChatQwen	1147	3.2K	14.2B	Tongyi Qianwen	12.2K	118
112	granite-3.1-8b-instructibm-granite	1145	211	8.2B	Apache 2.0	27.4K	169
113	Qwen1.5-7B-ChatQwen	1141	417	7.7B	Tongyi Qianwen	7K	192
114	granite-3.1-2b-instructibm-granite	1139	231	2.5B	Apache 2.0	24.5K	57
115	gemma-2-2b-itgoogle	1132	3.8K	2.6B	Gemma	125.7K	1,607
116	deepseek-llm-67b-chatdeepseek-ai	1132	211	—	Custom	2.8K	222
117	Mixtral-8x22B-Instruct-v0.1mistralai	1116	5.7K	141B	Apache 2.0	38.7K	780
118	Meta-Llama-3-70B-Instructmeta-llama	1114	15.9K	70.6B	Llama 3	37.8K	1,537
119	Phi-3-medium-4k-instructmicrosoft	1108	2.8K	14.0B	MIT	18.7K	231
120	Qwen-14B-ChatQwen	1077	195	14.2B	—	5.2K	381
121	Meta-Llama-3-8B-Instructmeta-llama	1076	10.8K	8.0B	Llama 3	359.6K	5,787
122	granite-3.0-2b-instructibm-granite	1070	759	2.6B	Apache 2.0	3.2K	48
123	granite-3.0-8b-instructibm-granite	1063	701	8.2B	Apache 2.0	13.3K	209
124	Phi-3-small-8k-instructmicrosoft	1061	2.4K	7.4B	MIT	7.1K	182
125	gemma-1.1-7b-itgoogle	1061	3.2K	8.5B	Gemma	4.2K	276
126	Mixtral-8x7B-Instruct-v0.1mistralai	1055	8.3K	46.7B	Apache 2.0	298.7K	4,777
127	gemma-7b-itgoogle	1036	1K	8.5B	Gemma	25.3K	1,263
128	Phi-3-mini-4k-instructmicrosoft	1033	1.2K	3.8B	MIT	329.2K	1,489
129	Qwen1.5-4B-ChatQwen	1024	1K	4.0B	Custom	15.9K	53
130	Llama-3.2-3B-Instructmeta-llama	1017	721	3.2B	Llama 3.2	914.1K	2,963
131	Phi-3-mini-128k-instructmicrosoft	1016	2K	3.8B	MIT	238.7K	1,713
132	OLMo-7B-Instructallenai	1014	722	6.9B	Apache 2.0	846	54
133	gemma-1.1-2b-itgoogle	1012	1.3K	2.5B	Gemma	6.1K	175
134	Mistral-7B-Instruct-v0.2mistralai	1009	2K	7.2B	Apache 2.0	1.6M	3,280
135	Llama-2-13b-chat-hfmeta-llama	1001	1.2K	13.0B	Llama 2	148.8K	1,138
136	Llama-2-70b-chat-hfmeta-llama	995	3.6K	69.0B	Llama 2	4.8K	2,211
137	gemma-2b-itgoogle	986	598	2.5B	Gemma	14K	975
138	Llama-2-7b-chat-hfmeta-llama	973	956	6.7B	Llama 2	126.4K	4,951
139	Llama-3.2-1B-Instructmeta-llama	959	674	1.2B	Llama 3.2	1.5M	1,961

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.