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 hard prompts 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	Kimi-K3moonshotai	1495	26.1K	2.8T	Custom	771.6K	11,950
2	GLM-5.3zai-org	1492	19.4K	753B	Custom	2.1M	2,395
3	GLM-5.3-Flashzai-org	1492	32K	321B	MIT	6.4M	3,088
4	MiMo-V2.5-ProXiaomiMiMo	1485	50.9K	1.0T	MIT	9.9K	771
5	GLM-5.2zai-org	1481	35.9K	753B	MIT	318.5K	5,205
6	GLM-5.1zai-org	1473	44.7K	754B	MIT	892.7K	1,850
7	Kimi-K2.6moonshotai	1471	26K	1.0T	Modified MIT	237.2K	1,641
8	DeepSeek-V4-Prodeepseek-ai	1461	38.4K	1.6T	MIT	335.6K	5,699
9	Qwen3.8-27BQwen	1456	21K	27.8B	Apache 2.0	7.8M	19,348
10	Kimi-K2.5moonshotai	1453	47.8K	1.0T	Modified MIT	395K	2,901
11	Inklingthinkingmachines	1453	32.8K	952B	Apache 2.0	1.8M	1,842
12	NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16nvidia	1452	7.6K	561B	OpenMDW 1.1	1.1M	384
13	GLM-5zai-org	1452	18.2K	754B	MIT	1.4M	2,141
14	MiMo-V2.5XiaomiMiMo	1450	31.8K	311B	MIT	110.3K	434
15	Hy3tencent	1448	10K	299B	Apache 2.0	1.2M	990
16	Qwen3.5-397B-A17BQwen	1448	65.5K	403B	Apache 2.0	857.2K	1,581
17	gemma-4-31Bgoogle	1447	3.6K	31.3B	Apache 2.0	57K	589
18	DeepSeek-V4-Flashdeepseek-ai	1444	34.7K	291B	MIT	1.1M	2,378
19	GLM-4.7zai-org	1443	6.8K	358B	MIT	66.6K	2,077
20	MiniMax-M3MiniMaxAI	1443	48.4K	427B	Custom	155.9K	1,600
21	Beam-1.0-501B-A23Breflection	1443	2.8K	501B	Apache 2.0	11.8K	311
22	GLM-4.6zai-org	1440	19.4K	357B	MIT	28.1K	1,247
23	gemma-4-26B-A4Bgoogle	1438	3.5K	25.8B	Apache 2.0	74.1K	465
24	Mistral-Medium-3.5-128Bmistralai	1436	7.6K	128B	Custom	17.2K	498
25	DeepSeek-V3.2deepseek-ai	1433	27.1K	685B	MIT	7.5M	1,596
26	Qwen3-235B-A22B-Instruct-2507Qwen	1433	53.5K	235B	Apache 2.0	84.8K	814
27	LongCat-Flash-Chatmeituan-longcat	1433	5.5K	562B	MIT	13.3K	540
28	Mistral-Large-3-675B-Instruct-2512mistralai	1430	55.2K	—	Apache 2.0	2.5K	261
29	GLM-4.5zai-org	1429	11.1K	358B	MIT	110.4K	1,439
30	DeepSeek-V3.2-Expdeepseek-ai	1429	6.6K	685B	MIT	43.6K	1,027
31	Qwen3-VL-235B-A22B-InstructQwen	1428	5.7K	236B	Apache 2.0	327K	431
32	Qwen3-Next-80B-A3B-InstructQwen	1428	12K	81.3B	Apache 2.0	129.9K	1,072
33	DeepSeek-V3.1-Terminusdeepseek-ai	1426	1.7K	685B	MIT	9.5K	386
34	Inkling-Smallthinkingmachines	1424	25.4K	266B	Apache 2.0	1.2M	449
35	MiniMax-M2.7MiniMaxAI	1422	59.1K	229B	Custom	307.1K	1,261
36	Qwen3.5-122B-A10BQwen	1421	18.8K	125B	Apache 2.0	924.6K	653
37	MiMo-V2-FlashXiaomiMiMo	1420	29.1K	310B	MIT	22.2K	756
38	DeepSeek-V3.1deepseek-ai	1417	5.1K	685B	MIT	262.8K	864
39	Qwen3-235B-A22B-Thinking-2507Qwen	1416	3.8K	235B	Apache 2.0	8.9K	419
40	DeepSeek-R1-0528deepseek-ai	1415	6.9K	685B	MIT	122.3K	2,495
41	Qwen3.5-27BQwen	1414	18K	27.8B	Apache 2.0	1.7M	1,090
42	Hy3-previewtencent	1412	4.6K	299B	Custom	68.3K	314
43	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16nvidia	1412	2.7K	124B	OpenMDW 1.1	70.3K	116
44	Step-3.5-Flashstepfun-ai	1411	37.5K	199B	Apache 2.0	219K	847
45	Qwen3-VL-235B-A22B-ThinkingQwen	1407	4K	236B	Apache 2.0	81K	420
46	Qwen3.5-35B-A3BQwen	1400	19.3K	36.0B	Apache 2.0	1.6M	1,555
47	Qwen3-30B-A3B-Instruct-2507Qwen	1398	11K	30.5B	Apache 2.0	932.4K	855
48	Muse-Glimmer-30Bmeta-models	1396	2.6K	29.8B	Apache 2.0	187.5K	2,127
49	Qwen3-235B-A22BQwen	1393	16.4K	235B	Apache 2.0	308.9K	1,128
50	NVIDIA-Nemotron-3-Super-120B-A12B-BF16nvidia	1386	4.4K	124B	Custom	1.1M	438
51	GLM-4.5-Airzai-org	1379	14.9K	110B	MIT	1.3M	649
52	granite-4.2-30bibm-granite	1374	2.1K	29.3B	Apache 2.0	32.9K	140
53	MiniMax-M2.5MiniMaxAI	1372	26.8K	229B	Modified MIT	1.2M	1,529
54	Qwen3-Coder-480B-A35B-InstructQwen	1372	11.5K	480B	Apache 2.0	58.6K	1,387
55	Qwen3-Next-80B-A3B-ThinkingQwen	1371	6.7K	81.3B	Apache 2.0	44.4K	503
56	GLM-4.6Vzai-org	1368	1.5K	108B	MIT	5.5K	398
57	DeepSeek-V3-0324deepseek-ai	1365	18.6K	685B	MIT	153.8K	3,223
58	gpt-oss-120bopenai	1363	14.7K	117B	Apache 2.0	2.4M	5,498
59	Ling-flash-2.0inclusionAI	1363	3.5K	103B	MIT	4.4K	222
60	DeepSeek-R1deepseek-ai	1361	4.1K	684B	MIT	1.6M	14,439
61	MiniMax-M2MiniMaxAI	1357	3.7K	229B	Modified MIT	109.4K	1,514
62	GLM-4.7-Flashzai-org	1356	6.7K	31.2B	MIT	1M	1,896
63	INTELLECT-3PrimeIntellect	1352	2.8K	107B	MIT	774	218
64	Trinity-Large-Previewarcee-ai	1349	19.5K	399B	OpenMDW 1.1	1.7K	181
65	NVIDIA-Nemotron-3-Nano-30B-A3B-BF16nvidia	1348	8.4K	31.6B	Custom	1.6M	840
66	Ring-flash-2.0inclusionAI	1344	3.5K	103B	MIT	1.4K	104
67	Trinity-Large-Thinkingarcee-ai	1342	20K	399B	OpenMDW 1.1	3.2K	190
68	NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16nvidia	1341	16.7K	31.6B	OpenMDW 1.1	1.1M	271
69	gemma-3-27b-itgoogle	1340	17.7K	27.4B	Gemma	364.5K	2,066
70	Mistral-Small-3.2-24B-Instruct-2506mistralai	1335	7.8K	24.0B	Apache 2.0	109.2K	644
71	GLM-4.5Vzai-org	1335	2.3K	108B	MIT	16.2K	726
72	Qwen3-32BQwen	1334	1.2K	32.8B	Apache 2.0	3M	791
73	granite-4.2-8bibm-granite	1326	2.1K	8.8B	Apache 2.0	69.2K	116
74	QwQ-32BQwen	1325	8.8K	32.8B	Apache 2.0	32K	2,971
75	Olmo-3.1-32B-Instructallenai	1322	6.3K	32.2B	Apache 2.0	20.8K	88
76	Qwen3-30B-A3BQwen	1314	10.6K	30.5B	Apache 2.0	1.1M	982
77	DeepSeek-V3deepseek-ai	1312	5.4K	684B	—	1.6M	4,462
78	gemma-3-12b-itgoogle	1309	977	12.2B	Gemma	365.8K	908
79	granite-4.2-3bibm-granite	1305	2K	3.7B	Apache 2.0	38.9K	131
80	Olmo-3-32B-Thinkallenai	1301	3K	32.2B	Apache 2.0	12.2K	177
81	granite-4.1-8bibm-granite	1293	2.7K	8.8B	Apache 2.0	98.1K	257
82	DeepSeek-V2.5-1210deepseek-ai	1289	1.8K	236B	Custom	2.1K	275
83	Molmo2-8Ballenai	1287	451	8.7B	Apache 2.0	44.4K	198
84	gemma-3n-E4B-itgoogle	1284	8.5K	7.8B	Gemma	12.7K	938
85	Llama-4-Maverick-17B-128E-Instructmeta-llama	1281	15.9K	402B	Llama 4	11.7K	520
86	Mistral-Small-3.1-24B-Instruct-2503mistralai	1278	14.6K	24.0B	Apache 2.0	168.4K	1,408
87	gpt-oss-20bopenai	1274	4.7K	20.9B	Apache 2.0	5.9M	5,286
88	Olmo-3.1-32B-Thinkallenai	1272	4.4K	32.2B	Apache 2.0	7K	120
89	Qwen2.5-72B-InstructQwen	1270	10.5K	72.7B	Custom	192.8K	1,043
90	DeepSeek-V2.5deepseek-ai	1270	6.9K	236B	Custom	4.5K	750
91	Llama-3.1-405B-Instructmeta-llama	1269	10.7K	406B	Llama 3.1	9.2K	607
92	Llama-3.1-Nemotron-70B-Instructnvidia	1266	2K	—	Llama 3.1	74	570
93	Llama-4-Scout-17B-16E-Instructmeta-llama	1266	12.9K	109B	Llama 4	110.7K	1,387
94	Llama-3.3-70B-Instructmeta-llama	1257	17K	70.6B	Llama 3.3	357.4K	3,197
95	gemma-3-4b-itgoogle	1252	1.1K	4.3B	Gemma	1M	1,633
96	Qwen2.5-Coder-32B-InstructQwen	1251	1.4K	32.8B	Apache 2.0	548.8K	2,206
97	Llama-3.1-70B-Instructmeta-llama	1241	15K	70.6B	Llama 3.1	92K	1,027
98	granite-4.0-h-smallibm-granite	1241	3K	32.2B	Apache 2.0	23.5K	313
99	Mistral-Small-24B-Instruct-2501mistralai	1233	3.6K	23.6B	Apache 2.0	94.4K	991
100	Llama-3.1-Tulu-3-70Ballenai	1220	779	70.6B	Llama 3.1	965	61
101	phi-4microsoft	1219	5.7K	14.7B	MIT	293.7K	2,340
102	DeepSeek-Coder-V2-Instructdeepseek-ai	1207	4.4K	236B	Custom	3.8K	718
103	Nemotron-4-340B-Instructnvidia	1201	5.6K	—	Custom	638	703
104	gemma-2-27b-itgoogle	1198	20.5K	27.2B	Gemma	21.7K	577
105	Meta-Llama-3-70B-Instructmeta-llama	1195	46K	70.6B	Llama 3	37.8K	1,537
106	aya-expanse-32bCohereLabs	1193	7.2K	32.3B	CC BY-NC 4.0	3K	303
107	Qwen2-72B-InstructQwen	1190	10.8K	72.7B	Tongyi Qianwen	24.3K	724
108	Llama-3.1-8B-Instructmeta-llama	1174	13.5K	8.0B	Llama 3.1	4.3M	8,836
109	Llama-3.1-Tulu-3-8Ballenai	1174	728	8.0B	Llama 3.1	5K	178
110	gemma-2-9b-itgoogle	1171	14.8K	9.2B	Gemma	113.6K	1,166
111	Qwen1.5-110B-ChatQwen	1168	7.7K	111B	Tongyi Qianwen	2.7K	139
112	QwQ-32B-PreviewQwen	1162	852	32.8B	Apache 2.0	7.1K	1,747
113	aya-expanse-8bCohereLabs	1156	2.6K	8.0B	CC BY-NC 4.0	22.9K	463
114	Mixtral-8x22B-Instruct-v0.1mistralai	1150	14.7K	141B	Apache 2.0	38.7K	780
115	Qwen1.5-72B-ChatQwen	1148	10.9K	72.3B	Tongyi Qianwen	11.2K	224
116	granite-3.1-8b-instructibm-granite	1145	798	8.2B	Apache 2.0	27.4K	169
117	granite-3.1-2b-instructibm-granite	1138	832	2.5B	Apache 2.0	24.5K	57
118	Meta-Llama-3-8B-Instructmeta-llama	1133	30.2K	8.0B	Llama 3	359.6K	5,787
119	Qwen1.5-32B-ChatQwen	1130	6.1K	32.5B	Tongyi Qianwen	46.8K	115
120	Phi-3-medium-4k-instructmicrosoft	1126	6.8K	14.0B	MIT	18.7K	231
121	Mixtral-8x7B-Instruct-v0.1mistralai	1114	19.5K	46.7B	Apache 2.0	298.7K	4,777
122	Qwen1.5-14B-ChatQwen	1113	4.9K	14.2B	Tongyi Qianwen	12.2K	118
123	gemma-2-2b-itgoogle	1112	12.3K	2.6B	Gemma	125.7K	1,607
124	Phi-3-small-8k-instructmicrosoft	1100	5.2K	7.4B	MIT	7.1K	182
125	Llama-3.2-3B-Instructmeta-llama	1094	2.2K	3.2B	Llama 3.2	914.1K	2,963
126	granite-3.0-8b-instructibm-granite	1091	1.7K	8.2B	Apache 2.0	13.3K	209
127	Phi-3-mini-4k-instructmicrosoft	1087	3.2K	3.8B	MIT	329.2K	1,489
128	granite-3.0-2b-instructibm-granite	1073	1.7K	2.6B	Apache 2.0	3.2K	48
129	Llama-2-70b-chat-hfmeta-llama	1073	9.7K	69.0B	Llama 2	4.8K	2,211
130	gemma-1.1-7b-itgoogle	1070	7K	8.5B	Gemma	4.2K	276
131	deepseek-llm-67b-chatdeepseek-ai	1070	1.1K	—	Custom	2.8K	222
132	Mistral-7B-Instruct-v0.2mistralai	1067	5.2K	7.2B	Apache 2.0	1.6M	3,280
133	Qwen1.5-7B-ChatQwen	1065	1.3K	7.7B	Tongyi Qianwen	7K	192
134	SOLAR-10.7B-Instruct-v1.0upstage	1062	932	10.7B	CC BY-NC 4.0	48.7K	660
135	Llama-2-13b-chat-hfmeta-llama	1051	4.5K	13.0B	Llama 2	148.8K	1,138
136	Llama-3.2-1B-Instructmeta-llama	1044	2.2K	1.2B	Llama 3.2	1.5M	1,961
137	gemma-7b-itgoogle	1041	2.3K	8.5B	Gemma	25.3K	1,263
138	Phi-3-mini-128k-instructmicrosoft	1028	6.2K	3.8B	MIT	238.7K	1,713
139	Qwen-14B-ChatQwen	1027	1K	14.2B	—	5.2K	381
140	Llama-2-7b-chat-hfmeta-llama	1009	3.3K	6.7B	Llama 2	126.4K	4,951
141	gemma-1.1-2b-itgoogle	1005	3.2K	2.5B	Gemma	6.1K	175
142	OLMo-7B-Instructallenai	993	1.5K	6.9B	Apache 2.0	846	54
143	gemma-2b-itgoogle	989	1.2K	2.5B	Gemma	14K	975
144	Qwen1.5-4B-ChatQwen	976	2.1K	4.0B	Custom	15.9K	53

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.