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	Code	Tools	Context	Speed	Runs
1	QQwen3.5-35B-A3BQ4_K_M	96%	9/9	8/9	9/9	201.4 tok/s	3
2	QQwen3.5-27BQ4_K_M	93%	8/9	9/9	8/9	35.5 tok/s	3
3	QQwen3-30B-A3B-Instruct-2507Q4_K_M	92%	11/12	10/12	12/12	187.6 tok/s	4
4	GGLM-4.7-FlashQ4_K_M	89%	9/9	6/9	9/9	201.7 tok/s	3
5	QQwen3-Coder-30B-A3B-InstructQ4_K_M	89%	9/9	7/9	8/9	186.4 tok/s	3
6	QQwen3-8BQ4_K_M	89%	9/12	11/12	12/12	120.7 tok/s	4
7	QQwen3-14BQ4_K_M	87%	13/15	11/15	15/15	67.1 tok/s	5
8	Ggemma-3-27b-itQ4_K_M	86%	17/18	Not supported	14/18	36.2 tok/s	6
9	Ggpt-oss-20bMXFP4	85%	9/9	9/9	5/9	173.8 tok/s	3
10	Ggemma-3-4b-itQ4_K_M	83%	11/15	Not supported	14/15	232.7 tok/s	5
11	QQwen3.5-2BQ4_K_M	81%	9/12	9/12	11/12	298.1 tok/s	4
12	QQwen3-4BQ4_K_M	78%	6/9	6/9	9/9	247.9 tok/s	3
13	Ggemma-3-12b-itQ4_K_M	75%	8/12	Not supported	10/12	80.9 tok/s	4
14	LLlama-3.1-8B-InstructQ4_K_M	71%	8/15	9/15	15/15	124.4 tok/s	5
15	PPhi-3.5-mini-instructQ4_K_M	64%	9/18	Not supported	14/18	261.3 tok/s	6
16	MMiniCPM5-2BQ4_K_M	52%	4/9	3/9	7/9	301.4 tok/s	3
17	MMistral-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 multi-turn ranked by LMArena score · updated Nov 6, 2026

#	Model	Score	Votes	Params	License	Downloads	Likes
1	KKimi-K3moonshotai	1489	6.3K	2.8T	Custom	771.6K	11,950
2	MMiMo-V2.5-ProXiaomiMiMo	1476	13.2K	1.0T	MIT	9.9K	771
3	GGLM-5.3zai-org	1473	4.6K	753B	Custom	2.1M	2,395
4	GGLM-5.1zai-org	1472	10.8K	754B	MIT	892.7K	1,850
5	GGLM-5.3-Flashzai-org	1471	7.7K	321B	MIT	6.4M	3,088
6	GGLM-5.2zai-org	1469	8.8K	753B	MIT	318.5K	5,205
7	DDeepSeek-V4-Prodeepseek-ai	1468	10.4K	1.6T	MIT	335.6K	5,699
8	GGLM-5zai-org	1456	4.7K	754B	MIT	1.4M	2,141
9	KKimi-K2.6moonshotai	1454	6.9K	1.0T	Modified MIT	237.2K	1,641
10	Ggemma-4-31Bgoogle	1451	1.1K	31.3B	Apache 2.0	57K	589
11	DDeepSeek-V4-Flashdeepseek-ai	1449	9.3K	291B	MIT	1.1M	2,378
12	GGLM-4.7zai-org	1446	1.9K	358B	MIT	66.6K	2,077
13	QQwen3.5-397B-A17BQwen	1446	16K	403B	Apache 2.0	857.2K	1,581
14	MMiMo-V2.5XiaomiMiMo	1445	8.4K	311B	MIT	110.3K	434
15	KKimi-K2.5moonshotai	1444	13.1K	1.0T	Modified MIT	395K	2,901
16	Ggemma-4-26B-A4Bgoogle	1442	1.1K	25.8B	Apache 2.0	74.1K	465
17	IInklingthinkingmachines	1441	7.9K	952B	Apache 2.0	1.8M	1,842
18	MMiniMax-M3MiniMaxAI	1439	12.7K	427B	Custom	155.9K	1,600
19	QQwen3.8-27BQwen	1436	5.1K	27.8B	Apache 2.0	7.8M	19,348
20	HHy3tencent	1436	2.5K	299B	Apache 2.0	1.2M	990
21	QQwen3-235B-A22B-Instruct-2507Qwen	1432	17.1K	235B	Apache 2.0	84.8K	814
22	MMistral-Large-3-675B-Instruct-2512mistralai	1430	15K	—	Apache 2.0	2.5K	261
23	QQwen3-VL-235B-A22B-InstructQwen	1428	1.9K	236B	Apache 2.0	327K	431
24	DDeepSeek-V3.2deepseek-ai	1427	8.3K	685B	MIT	7.5M	1,596
25	GGLM-4.6zai-org	1427	5.6K	357B	MIT	28.1K	1,247
26	MMistral-Medium-3.5-128Bmistralai	1423	1.9K	128B	Custom	17.2K	498
27	DDeepSeek-V3.2-Expdeepseek-ai	1421	2K	685B	MIT	43.6K	1,027
28	BBeam-1.0-501B-A23Breflection	1419	707	501B	Apache 2.0	11.8K	311
29	QQwen3.5-122B-A10BQwen	1417	5.2K	125B	Apache 2.0	924.6K	653
30	IInkling-Smallthinkingmachines	1416	6K	266B	Apache 2.0	1.2M	449
31	QQwen3-Next-80B-A3B-InstructQwen	1416	4K	81.3B	Apache 2.0	129.9K	1,072
32	GGLM-4.5zai-org	1415	3.7K	358B	MIT	110.4K	1,439
33	MMiniMax-M2.7MiniMaxAI	1412	15.5K	229B	Custom	307.1K	1,261
34	NNVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16nvidia	1412	669	124B	OpenMDW 1.1	70.3K	116
35	NNVIDIA-Nemotron-3-Ultra-550B-A55B-BF16nvidia	1411	1.8K	561B	OpenMDW 1.1	1.1M	384
36	DDeepSeek-V3.1-Terminusdeepseek-ai	1411	572	685B	MIT	9.5K	386
37	HHy3-previewtencent	1410	1.2K	299B	Custom	68.3K	314
38	QQwen3.5-27BQwen	1410	4.9K	27.8B	Apache 2.0	1.7M	1,090
39	DDeepSeek-V3.1deepseek-ai	1408	1.8K	685B	MIT	262.8K	864
40	QQwen3-235B-A22B-Thinking-2507Qwen	1406	1.4K	235B	Apache 2.0	8.9K	419
41	SStep-3.5-Flashstepfun-ai	1405	10.4K	199B	Apache 2.0	219K	847
42	DDeepSeek-R1-0528deepseek-ai	1404	2.8K	685B	MIT	122.3K	2,495
43	MMiMo-V2-FlashXiaomiMiMo	1404	8.5K	310B	MIT	22.2K	756
44	LLongCat-Flash-Chatmeituan-longcat	1403	2K	562B	MIT	13.3K	540
45	QQwen3-235B-A22BQwen	1398	6.5K	235B	Apache 2.0	308.9K	1,128
46	MMuse-Glimmer-30Bmeta-models	1398	619	29.8B	Apache 2.0	187.5K	2,127
47	DDeepSeek-R1deepseek-ai	1390	2.4K	684B	MIT	1.6M	14,439
48	QQwen3.5-35B-A3BQwen	1390	5.5K	36.0B	Apache 2.0	1.6M	1,555
49	DDeepSeek-V3-0324deepseek-ai	1389	7.7K	685B	MIT	153.8K	3,223
50	QQwen3-VL-235B-A22B-ThinkingQwen	1387	1.3K	236B	Apache 2.0	81K	420
51	QQwen3-30B-A3B-Instruct-2507Qwen	1379	3.9K	30.5B	Apache 2.0	932.4K	855
52	GGLM-4.5-Airzai-org	1371	5K	110B	MIT	1.3M	649
53	NNVIDIA-Nemotron-3-Super-120B-A12B-BF16nvidia	1369	1.4K	124B	Custom	1.1M	438
54	QQwen3-Coder-480B-A35B-InstructQwen	1365	4.3K	480B	Apache 2.0	58.6K	1,387
55	MMiniMax-M2.5MiniMaxAI	1364	7.7K	229B	Modified MIT	1.2M	1,529
56	MMiniMax-M2MiniMaxAI	1361	1.1K	229B	Modified MIT	109.4K	1,514
57	GGLM-4.6Vzai-org	1358	490	108B	MIT	5.5K	398
58	IINTELLECT-3PrimeIntellect	1351	794	107B	MIT	774	218
59	DDeepSeek-V3deepseek-ai	1349	3.7K	684B	—	1.6M	4,462
60	QQwen3-Next-80B-A3B-ThinkingQwen	1346	2.2K	81.3B	Apache 2.0	44.4K	503
61	Ggemma-3-27b-itgoogle	1345	6.7K	27.4B	Gemma	364.5K	2,066
62	MMistral-Small-3.2-24B-Instruct-2506mistralai	1343	3K	24.0B	Apache 2.0	109.2K	644
63	GGLM-4.7-Flashzai-org	1342	2.1K	31.2B	MIT	1M	1,896
64	TTrinity-Large-Previewarcee-ai	1342	5.6K	399B	OpenMDW 1.1	1.7K	181
65	Ggpt-oss-120bopenai	1340	4.9K	117B	Apache 2.0	2.4M	5,498
66	Ggranite-4.2-30bibm-granite	1340	547	29.3B	Apache 2.0	32.9K	140
67	Ggemma-3-12b-itgoogle	1334	471	12.2B	Gemma	365.8K	908
68	GGLM-4.5Vzai-org	1333	850	108B	MIT	16.2K	726
69	QQwen3-32BQwen	1331	590	32.8B	Apache 2.0	3M	791
70	NNVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16nvidia	1325	3.6K	31.6B	OpenMDW 1.1	1.1M	271
71	LLing-flash-2.0inclusionAI	1323	1.1K	103B	MIT	4.4K	222
72	TTrinity-Large-Thinkingarcee-ai	1320	5.6K	399B	OpenMDW 1.1	3.2K	190
73	NNVIDIA-Nemotron-3-Nano-30B-A3B-BF16nvidia	1320	2.6K	31.6B	Custom	1.6M	840
74	QQwQ-32BQwen	1314	3.8K	32.8B	Apache 2.0	32K	2,971
75	OOlmo-3.1-32B-Instructallenai	1310	2.1K	32.2B	Apache 2.0	20.8K	88
76	QQwen3-30B-A3BQwen	1307	4.4K	30.5B	Apache 2.0	1.1M	982
77	Ggranite-4.2-8bibm-granite	1300	520	8.8B	Apache 2.0	69.2K	116
78	LLlama-3.1-405B-Instructmeta-llama	1297	6.8K	406B	Llama 3.1	9.2K	607
79	DDeepSeek-V2.5-1210deepseek-ai	1297	1.2K	236B	Custom	2.1K	275
80	Ggranite-4.2-3bibm-granite	1292	528	3.7B	Apache 2.0	38.9K	131
81	OOlmo-3-32B-Thinkallenai	1290	771	32.2B	Apache 2.0	12.2K	177
82	LLlama-4-Maverick-17B-128E-Instructmeta-llama	1289	6.7K	402B	Llama 4	11.7K	520
83	RRing-flash-2.0inclusionAI	1288	1.1K	103B	MIT	1.4K	104
84	LLlama-4-Scout-17B-16E-Instructmeta-llama	1280	5.2K	109B	Llama 4	110.7K	1,387
85	LLlama-3.3-70B-Instructmeta-llama	1279	9.1K	70.6B	Llama 3.3	357.4K	3,197
86	Ggemma-3n-E4B-itgoogle	1276	3.3K	7.8B	Gemma	12.7K	938
87	LLlama-3.1-Nemotron-70B-Instructnvidia	1275	1.2K	—	Llama 3.1	74	570
88	QQwen2.5-72B-InstructQwen	1272	7.3K	72.7B	Custom	192.8K	1,043
89	MMistral-Small-3.1-24B-Instruct-2503mistralai	1270	5.4K	24.0B	Apache 2.0	168.4K	1,408
90	Ggpt-oss-20bopenai	1268	1.8K	20.9B	Apache 2.0	5.9M	5,286
91	Ggranite-4.1-8bibm-granite	1268	703	8.8B	Apache 2.0	98.1K	257
92	DDeepSeek-V2.5deepseek-ai	1261	4.5K	236B	Custom	4.5K	750
93	LLlama-3.1-70B-Instructmeta-llama	1256	10.4K	70.6B	Llama 3.1	92K	1,027
94	Ggemma-3-4b-itgoogle	1255	565	4.3B	Gemma	1M	1,633
95	OOlmo-3.1-32B-Thinkallenai	1252	1.2K	32.2B	Apache 2.0	7K	120
96	LLlama-3.1-Tulu-3-70Ballenai	1251	537	70.6B	Llama 3.1	965	61
97	Ggranite-4.0-h-smallibm-granite	1242	944	32.2B	Apache 2.0	23.5K	313
98	Ggemma-2-27b-itgoogle	1224	13.6K	27.2B	Gemma	21.7K	577
99	MMeta-Llama-3-70B-Instructmeta-llama	1223	21.4K	70.6B	Llama 3	37.8K	1,537
100	QQwen2.5-Coder-32B-InstructQwen	1222	1K	32.8B	Apache 2.0	548.8K	2,206
101	MMistral-Small-24B-Instruct-2501mistralai	1217	2.2K	23.6B	Apache 2.0	94.4K	991
102	NNemotron-4-340B-Instructnvidia	1214	3.1K	—	Custom	638	703
103	Pphi-4microsoft	1206	3.5K	14.7B	MIT	293.7K	2,340
104	QQwen2-72B-InstructQwen	1195	6.7K	72.7B	Tongyi Qianwen	24.3K	724
105	Ggemma-2-9b-itgoogle	1193	9.7K	9.2B	Gemma	113.6K	1,166
106	Aaya-expanse-32bCohereLabs	1190	5.2K	32.3B	CC BY-NC 4.0	3K	303
107	DDeepSeek-Coder-V2-Instructdeepseek-ai	1176	2.7K	236B	Custom	3.8K	718
108	LLlama-3.1-8B-Instructmeta-llama	1172	8.9K	8.0B	Llama 3.1	4.3M	8,836
109	QQwen1.5-72B-ChatQwen	1160	5.5K	72.3B	Tongyi Qianwen	11.2K	224
110	QQwen1.5-110B-ChatQwen	1160	3.7K	111B	Tongyi Qianwen	2.7K	139
111	Aaya-expanse-8bCohereLabs	1160	1.7K	8.0B	CC BY-NC 4.0	22.9K	463
112	LLlama-3.1-Tulu-3-8Ballenai	1154	520	8.0B	Llama 3.1	5K	178
113	MMeta-Llama-3-8B-Instructmeta-llama	1152	14.8K	8.0B	Llama 3	359.6K	5,787
114	QQwQ-32B-PreviewQwen	1141	487	32.8B	Apache 2.0	7.1K	1,747
115	QQwen1.5-32B-ChatQwen	1140	3.1K	32.5B	Tongyi Qianwen	46.8K	115
116	MMixtral-8x22B-Instruct-v0.1mistralai	1130	7.8K	141B	Apache 2.0	38.7K	780
117	Ggemma-2-2b-itgoogle	1118	7.9K	2.6B	Gemma	125.7K	1,607
118	MMixtral-8x7B-Instruct-v0.1mistralai	1115	10.9K	46.7B	Apache 2.0	298.7K	4,777
119	QQwen1.5-14B-ChatQwen	1110	2.6K	14.2B	Tongyi Qianwen	12.2K	118
120	Ggranite-3.1-8b-instructibm-granite	1108	551	8.2B	Apache 2.0	27.4K	169
121	LLlama-3.2-3B-Instructmeta-llama	1105	1.3K	3.2B	Llama 3.2	914.1K	2,963
122	Ggranite-3.1-2b-instructibm-granite	1099	579	2.5B	Apache 2.0	24.5K	57
123	LLlama-2-70b-chat-hfmeta-llama	1088	5.5K	69.0B	Llama 2	4.8K	2,211
124	PPhi-3-medium-4k-instructmicrosoft	1088	3.8K	14.0B	MIT	18.7K	231
125	Ddeepseek-llm-67b-chatdeepseek-ai	1082	620	—	Custom	2.8K	222
126	PPhi-3-small-8k-instructmicrosoft	1068	3.1K	7.4B	MIT	7.1K	182
127	Ggranite-3.0-8b-instructibm-granite	1063	1.1K	8.2B	Apache 2.0	13.3K	209
128	MMistral-7B-Instruct-v0.2mistralai	1062	2.6K	7.2B	Apache 2.0	1.6M	3,280
129	QQwen1.5-7B-ChatQwen	1061	589	7.7B	Tongyi Qianwen	7K	192
130	Ggranite-3.0-2b-instructibm-granite	1053	1.2K	2.6B	Apache 2.0	3.2K	48
131	LLlama-2-13b-chat-hfmeta-llama	1050	2.6K	13.0B	Llama 2	148.8K	1,138
132	PPhi-3-mini-4k-instructmicrosoft	1049	1.8K	3.8B	MIT	329.2K	1,489
133	SSOLAR-10.7B-Instruct-v1.0upstage	1045	601	10.7B	CC BY-NC 4.0	48.7K	660
134	Ggemma-1.1-7b-itgoogle	1040	3.6K	8.5B	Gemma	4.2K	276
135	LLlama-3.2-1B-Instructmeta-llama	1030	1.2K	1.2B	Llama 3.2	1.5M	1,961
136	LLlama-2-7b-chat-hfmeta-llama	1029	1.9K	6.7B	Llama 2	126.4K	4,951
137	QQwen-14B-ChatQwen	1022	753	14.2B	—	5.2K	381
138	OOLMo-7B-Instructallenai	1007	537	6.9B	Apache 2.0	846	54
139	PPhi-3-mini-128k-instructmicrosoft	989	2.5K	3.8B	MIT	238.7K	1,713
140	QQwen1.5-4B-ChatQwen	977	1.1K	4.0B	Custom	15.9K	53
141	Ggemma-7b-itgoogle	963	1.3K	8.5B	Gemma	25.3K	1,263
142	Ggemma-1.1-2b-itgoogle	958	1.5K	2.5B	Gemma	6.1K	175
143	Ggemma-2b-itgoogle	945	768	2.5B	Gemma	14K	975

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