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

143 open-weight models in coding 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-5.3-Flashzai-org	1511	12.8K	321B	MIT	6.4M	3,088
2	Kimi-K3moonshotai	1507	10.2K	2.8T	Custom	771.6K	11,950
3	MiMo-V2.5-ProXiaomiMiMo	1501	21K	1.0T	MIT	9.9K	771
4	GLM-5.3zai-org	1498	7.3K	753B	Custom	2.1M	2,395
5	Kimi-K2.6moonshotai	1489	11.1K	1.0T	Modified MIT	237.2K	1,641
6	GLM-5.2zai-org	1488	14.7K	753B	MIT	318.5K	5,205
7	GLM-5.1zai-org	1486	18.4K	754B	MIT	892.7K	1,850
8	Kimi-K2.5moonshotai	1484	1.9K	1.0T	Modified MIT	395K	2,901
9	Qwen3.8-27BQwen	1478	8.7K	27.8B	Apache 2.0	7.8M	19,348
10	Beam-1.0-501B-A23Breflection	1477	1.1K	501B	Apache 2.0	11.8K	311
11	DeepSeek-V4-Prodeepseek-ai	1469	17.1K	1.6T	MIT	335.6K	5,699
12	MiMo-V2.5XiaomiMiMo	1469	13.6K	311B	MIT	110.3K	434
13	MiniMax-M3MiniMaxAI	1468	20.6K	427B	Custom	155.9K	1,600
14	NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16nvidia	1467	3.2K	561B	OpenMDW 1.1	1.1M	384
15	Inklingthinkingmachines	1467	13.5K	952B	Apache 2.0	1.8M	1,842
16	Hy3tencent	1467	4.1K	299B	Apache 2.0	1.2M	990
17	LongCat-Flash-Chatmeituan-longcat	1466	2.2K	562B	MIT	13.3K	540
18	Qwen3.5-397B-A17BQwen	1464	28K	403B	Apache 2.0	857.2K	1,581
19	GLM-5zai-org	1461	7.6K	754B	MIT	1.4M	2,141
20	Mistral-Medium-3.5-128Bmistralai	1460	3.3K	128B	Custom	17.2K	498
21	gemma-4-31Bgoogle	1459	1.4K	31.3B	Apache 2.0	57K	589
22	DeepSeek-V4-Flashdeepseek-ai	1457	15.3K	291B	MIT	1.1M	2,378
23	GLM-4.7zai-org	1454	2.5K	358B	MIT	66.6K	2,077
24	MiniMax-M2.7MiniMaxAI	1454	25.7K	229B	Custom	307.1K	1,261
25	DeepSeek-V3.2deepseek-ai	1453	8.9K	685B	MIT	7.5M	1,596
26	Inkling-Smallthinkingmachines	1451	10.6K	266B	Apache 2.0	1.2M	449
27	GLM-4.6zai-org	1449	7.6K	357B	MIT	28.1K	1,247
28	Mistral-Large-3-675B-Instruct-2512mistralai	1448	22.6K	—	Apache 2.0	2.5K	261
29	gemma-4-26B-A4Bgoogle	1446	1.4K	25.8B	Apache 2.0	74.1K	465
30	Qwen3-235B-A22B-Instruct-2507Qwen	1445	21.7K	235B	Apache 2.0	84.8K	814
31	MiMo-V2-FlashXiaomiMiMo	1442	12.4K	310B	MIT	22.2K	756
32	Qwen3-Next-80B-A3B-InstructQwen	1440	4.8K	81.3B	Apache 2.0	129.9K	1,072
33	DeepSeek-V3.2-Expdeepseek-ai	1438	1.9K	685B	MIT	43.6K	1,027
34	Qwen3-VL-235B-A22B-InstructQwen	1438	2.3K	236B	Apache 2.0	327K	431
35	Step-3.5-Flashstepfun-ai	1436	15.2K	199B	Apache 2.0	219K	847
36	Qwen3.5-122B-A10BQwen	1436	8.2K	125B	Apache 2.0	924.6K	653
37	GLM-4.5zai-org	1434	4.7K	358B	MIT	110.4K	1,439
38	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16nvidia	1428	1K	124B	OpenMDW 1.1	70.3K	116
39	DeepSeek-R1-0528deepseek-ai	1427	2.7K	685B	MIT	122.3K	2,495
40	Qwen3.5-27BQwen	1426	7.9K	27.8B	Apache 2.0	1.7M	1,090
41	Qwen3-VL-235B-A22B-ThinkingQwen	1425	1.6K	236B	Apache 2.0	81K	420
42	DeepSeek-V3.1-Terminusdeepseek-ai	1425	632	685B	MIT	9.5K	386
43	Qwen3-235B-A22B-Thinking-2507Qwen	1424	1.6K	235B	Apache 2.0	8.9K	419
44	Hy3-previewtencent	1421	2K	299B	Custom	68.3K	314
45	DeepSeek-V3.1deepseek-ai	1417	1.9K	685B	MIT	262.8K	864
46	Muse-Glimmer-30Bmeta-models	1417	1.1K	29.8B	Apache 2.0	187.5K	2,127
47	Qwen3-30B-A3B-Instruct-2507Qwen	1416	4.6K	30.5B	Apache 2.0	932.4K	855
48	Qwen3-Coder-480B-A35B-InstructQwen	1412	4.8K	480B	Apache 2.0	58.6K	1,387
49	Qwen3.5-35B-A3BQwen	1410	8.4K	36.0B	Apache 2.0	1.6M	1,555
50	NVIDIA-Nemotron-3-Super-120B-A12B-BF16nvidia	1403	1.9K	124B	Custom	1.1M	438
51	Qwen3-235B-A22BQwen	1397	6.9K	235B	Apache 2.0	308.9K	1,128
52	GLM-4.5-Airzai-org	1397	6.1K	110B	MIT	1.3M	649
53	granite-4.2-30bibm-granite	1396	845	29.3B	Apache 2.0	32.9K	140
54	Qwen3-Next-80B-A3B-ThinkingQwen	1391	2.7K	81.3B	Apache 2.0	44.4K	503
55	GLM-4.6Vzai-org	1390	542	108B	MIT	5.5K	398
56	Ling-flash-2.0inclusionAI	1389	1.5K	103B	MIT	4.4K	222
57	NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16nvidia	1382	6.5K	31.6B	OpenMDW 1.1	1.1M	271
58	GLM-4.7-Flashzai-org	1382	2.8K	31.2B	MIT	1M	1,896
59	Trinity-Large-Previewarcee-ai	1381	8K	399B	OpenMDW 1.1	1.7K	181
60	MiniMax-M2.5MiniMaxAI	1380	11.3K	229B	Modified MIT	1.2M	1,529
61	gpt-oss-120bopenai	1379	6.4K	117B	Apache 2.0	2.4M	5,498
62	NVIDIA-Nemotron-3-Nano-30B-A3B-BF16nvidia	1379	3.3K	31.6B	Custom	1.6M	840
63	granite-4.2-8bibm-granite	1376	820	8.8B	Apache 2.0	69.2K	116
64	INTELLECT-3PrimeIntellect	1375	1K	107B	MIT	774	218
65	DeepSeek-R1deepseek-ai	1371	2.3K	684B	MIT	1.6M	14,439
66	MiniMax-M2MiniMaxAI	1371	1.5K	229B	Modified MIT	109.4K	1,514
67	DeepSeek-V3-0324deepseek-ai	1368	8.3K	685B	MIT	153.8K	3,223
68	Ring-flash-2.0inclusionAI	1366	1.5K	103B	MIT	1.4K	104
69	Mistral-Small-3.2-24B-Instruct-2506mistralai	1362	3.3K	24.0B	Apache 2.0	109.2K	644
70	granite-4.2-3bibm-granite	1360	824	3.7B	Apache 2.0	38.9K	131
71	Qwen3-32BQwen	1357	513	32.8B	Apache 2.0	3M	791
72	Trinity-Large-Thinkingarcee-ai	1357	8.5K	399B	OpenMDW 1.1	3.2K	190
73	GLM-4.5Vzai-org	1347	965	108B	MIT	16.2K	726
74	Olmo-3.1-32B-Instructallenai	1347	2.5K	32.2B	Apache 2.0	20.8K	88
75	Qwen3-30B-A3BQwen	1337	4.5K	30.5B	Apache 2.0	1.1M	982
76	QwQ-32BQwen	1333	4K	32.8B	Apache 2.0	32K	2,971
77	DeepSeek-V3deepseek-ai	1325	3.3K	684B	—	1.6M	4,462
78	gemma-3-27b-itgoogle	1322	8K	27.4B	Gemma	364.5K	2,066
79	Olmo-3-32B-Thinkallenai	1317	1K	32.2B	Apache 2.0	12.2K	177
80	granite-4.1-8bibm-granite	1312	1.1K	8.8B	Apache 2.0	98.1K	257
81	DeepSeek-V2.5-1210deepseek-ai	1309	1.1K	236B	Custom	2.1K	275
82	Mistral-Small-3.1-24B-Instruct-2503mistralai	1309	6.1K	24.0B	Apache 2.0	168.4K	1,408
83	gpt-oss-20bopenai	1306	2.1K	20.9B	Apache 2.0	5.9M	5,286
84	Llama-4-Maverick-17B-128E-Instructmeta-llama	1301	6.9K	402B	Llama 4	11.7K	520
85	DeepSeek-V2.5deepseek-ai	1301	4.3K	236B	Custom	4.5K	750
86	Qwen2.5-72B-InstructQwen	1292	6.7K	72.7B	Custom	192.8K	1,043
87	Olmo-3.1-32B-Thinkallenai	1292	1.6K	32.2B	Apache 2.0	7K	120
88	Llama-3.1-405B-Instructmeta-llama	1291	6.2K	406B	Llama 3.1	9.2K	607
89	Llama-4-Scout-17B-16E-Instructmeta-llama	1286	5.2K	109B	Llama 4	110.7K	1,387
90	gemma-3-12b-itgoogle	1281	543	12.2B	Gemma	365.8K	908
91	Qwen2.5-Coder-32B-InstructQwen	1275	873	32.8B	Apache 2.0	548.8K	2,206
92	Llama-3.1-Nemotron-70B-Instructnvidia	1272	1.3K	—	Llama 3.1	74	570
93	gemma-3n-E4B-itgoogle	1268	3.5K	7.8B	Gemma	12.7K	938
94	Llama-3.3-70B-Instructmeta-llama	1268	8.7K	70.6B	Llama 3.3	357.4K	3,197
95	Llama-3.1-70B-Instructmeta-llama	1259	9.4K	70.6B	Llama 3.1	92K	1,027
96	DeepSeek-Coder-V2-Instructdeepseek-ai	1251	2.7K	236B	Custom	3.8K	718
97	granite-4.0-h-smallibm-granite	1249	1.3K	32.2B	Apache 2.0	23.5K	313
98	Mistral-Small-24B-Instruct-2501mistralai	1245	2.1K	23.6B	Apache 2.0	94.4K	991
99	Llama-3.1-Tulu-3-70Ballenai	1235	450	70.6B	Llama 3.1	965	61
100	phi-4microsoft	1231	3.3K	14.7B	MIT	293.7K	2,340
101	gemma-3-4b-itgoogle	1230	605	4.3B	Gemma	1M	1,633
102	gemma-2-27b-itgoogle	1211	12.1K	27.2B	Gemma	21.7K	577
103	Nemotron-4-340B-Instructnvidia	1209	3.3K	—	Custom	638	703
104	Meta-Llama-3-70B-Instructmeta-llama	1206	28.1K	70.6B	Llama 3	37.8K	1,537
105	aya-expanse-32bCohereLabs	1196	4.7K	32.3B	CC BY-NC 4.0	3K	303
106	Qwen2-72B-InstructQwen	1196	6.2K	72.7B	Tongyi Qianwen	24.3K	724
107	Llama-3.1-8B-Instructmeta-llama	1195	8.6K	8.0B	Llama 3.1	4.3M	8,836
108	granite-3.1-8b-instructibm-granite	1186	478	8.2B	Apache 2.0	27.4K	169
109	Qwen1.5-110B-ChatQwen	1184	4.8K	111B	Tongyi Qianwen	2.7K	139
110	Llama-3.1-Tulu-3-8Ballenai	1183	476	8.0B	Llama 3.1	5K	178
111	gemma-2-9b-itgoogle	1173	8.9K	9.2B	Gemma	113.6K	1,166
112	Mixtral-8x22B-Instruct-v0.1mistralai	1165	8.8K	141B	Apache 2.0	38.7K	780
113	Qwen1.5-72B-ChatQwen	1165	6.4K	72.3B	Tongyi Qianwen	11.2K	224
114	aya-expanse-8bCohereLabs	1160	1.6K	8.0B	CC BY-NC 4.0	22.9K	463
115	QwQ-32B-PreviewQwen	1155	566	32.8B	Apache 2.0	7.1K	1,747
116	Qwen1.5-32B-ChatQwen	1154	3.9K	32.5B	Tongyi Qianwen	46.8K	115
117	Meta-Llama-3-8B-Instructmeta-llama	1151	18.4K	8.0B	Llama 3	359.6K	5,787
118	granite-3.1-2b-instructibm-granite	1149	508	2.5B	Apache 2.0	24.5K	57
119	Qwen1.5-14B-ChatQwen	1137	3.2K	14.2B	Tongyi Qianwen	12.2K	118
120	Phi-3-medium-4k-instructmicrosoft	1130	4K	14.0B	MIT	18.7K	231
121	Mixtral-8x7B-Instruct-v0.1mistralai	1126	11.8K	46.7B	Apache 2.0	298.7K	4,777
122	granite-3.0-8b-instructibm-granite	1112	1.1K	8.2B	Apache 2.0	13.3K	209
123	gemma-2-2b-itgoogle	1112	7.3K	2.6B	Gemma	125.7K	1,607
124	Qwen1.5-7B-ChatQwen	1107	772	7.7B	Tongyi Qianwen	7K	192
125	Phi-3-small-8k-instructmicrosoft	1101	3.2K	7.4B	MIT	7.1K	182
126	Llama-3.2-3B-Instructmeta-llama	1097	1.4K	3.2B	Llama 3.2	914.1K	2,963
127	deepseek-llm-67b-chatdeepseek-ai	1095	649	—	Custom	2.8K	222
128	Phi-3-mini-4k-instructmicrosoft	1093	1.8K	3.8B	MIT	329.2K	1,489
129	granite-3.0-2b-instructibm-granite	1090	1.1K	2.6B	Apache 2.0	3.2K	48
130	gemma-1.1-7b-itgoogle	1083	4.3K	8.5B	Gemma	4.2K	276
131	Mistral-7B-Instruct-v0.2mistralai	1081	3.1K	7.2B	Apache 2.0	1.6M	3,280
132	Llama-2-70b-chat-hfmeta-llama	1079	5.7K	69.0B	Llama 2	4.8K	2,211
133	Qwen-14B-ChatQwen	1071	599	14.2B	—	5.2K	381
134	Llama-3.2-1B-Instructmeta-llama	1070	1.3K	1.2B	Llama 3.2	1.5M	1,961
135	SOLAR-10.7B-Instruct-v1.0upstage	1067	482	10.7B	CC BY-NC 4.0	48.7K	660
136	Llama-2-13b-chat-hfmeta-llama	1062	2.6K	13.0B	Llama 2	148.8K	1,138
137	gemma-7b-itgoogle	1048	1.4K	8.5B	Gemma	25.3K	1,263
138	Phi-3-mini-128k-instructmicrosoft	1039	3.9K	3.8B	MIT	238.7K	1,713
139	gemma-1.1-2b-itgoogle	1034	2K	2.5B	Gemma	6.1K	175
140	OLMo-7B-Instructallenai	1015	772	6.9B	Apache 2.0	846	54
141	gemma-2b-itgoogle	1009	742	2.5B	Gemma	14K	975
142	Llama-2-7b-chat-hfmeta-llama	1002	2K	6.7B	Llama 2	126.4K	4,951
143	Qwen1.5-4B-ChatQwen	999	1.3K	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.