Models
Topics
Language
More

More filters

Show inactive models

Hide models that are no longer actively available on Yupp.

Turns

Filter model performance by the number of turns in a conversation.

Open license models

Filter the leaderboard to only show models that have an open license.

1270
Qwen3 Next 80B A3B Instruct
1269
gpt-oss-120b
1175
Kimi K2.5
1171
Kimi K2 Thinking Turbo
1163
Step 3.5 Flash
1158
DeepSeek V3.2 Thinking
1119
MiniMax M2.5 Lightning
1085
gpt-oss-20b
1073
DeepSeek-R1 Turbo
1069
Kimi K2 Fast
1061
Qwen3 32B Fast
1059
Nemotron 3 Nano (Thinking)
1054
DeepSeek V3.2 Exp Chat
1053
Qwen3.5 122B A17B
1053
Qwen3 14B

Last updated about 1 month ago

RankOverallNameVIBE
Score
Confidence
Interval
VotesDownvote %Abort %SpeedLatencyContextCost
(Input)
Cost
(Output)
133Qwen3 Next 80B A3B Instruct1270±102.3K2.8%0.6%84 tps1.1s256K$0.20$1.42
248gpt-oss-120b1269±74.6K1.4%0.7%213 tps0.5s131K$0.11$0.50
333Kimi K2.51175±123.1K1.6%6.5%33 tps1.7s262K$0.34$2.57
444Kimi K2 Thinking Turbo1171±111.8K4.2%2.0%75 tps1.4s262K$1.15$8.00
548Step 3.5 Flash1163±236400.8%2.2%109 tps0.6s256K$0.05$0.15
656DeepSeek V3.2 Thinking1158±192.4K2.3%9.0%30 tps2.6s131K$0.28$0.42
779MiniMax M2.5 Lightning1119±166500.8%1.5%51 tps2.0s205K$0.60$2.40
8101gpt-oss-20b1085±122.1K2.6%0.5%216 tps0.5s131K$0.06$0.26
995DeepSeek-R1 Turbo1073±167803.7%2.6%29 tps1.8s64K$2.85$4.75
10113Kimi K2 Fast1069±78.5K1.0%0.8%365 tps0.5s131K$1.00$3.00
11121Qwen3 32B Fast1061±93K2.4%11.6%30 tps3.1s41K$0.10$0.25
1286Nemotron 3 Nano (Thinking)1059±188251.8%2.0%200 tps0.5s256K$0$0
1365DeepSeek V3.2 Exp Chat1054±141.2K3.7%2.6%29 tps1.5s131K$0.27$0.39
1452Qwen3.5 122B A17B1053±255803.3%1.5%82 tps1.4s256K$0.40$3.20
15133Qwen3 14B1053±131.7K2.9%1.7%109 tps0.8s41K$0.04$0.15
1637Kimi K2.5 Instant1046±166202.4%2.9%32 tps3.0s262K$0.50$3.00
1795DeepSeek V3.2 Exp Thinking1046±187354.5%7.2%26 tps3.0s131K$0.28$0.42
18133DeepSeek-R1 05281038±131.7K2.0%1.3%93 tps0.5s64K$1.60$3.67
19121QwQ 32B1015±74.6K1.4%5.4%41 tps2.1s16K$0.43$0.56
20153Qwen 2.5 32B Instruct1004±141.2K1.7%2.5%48 tps1.0s131K$0.21$0.25
2186Qwen3 235B A22B998±181.4K3.2%5.3%71 tps0.9s41K$0.23$0.63
22126Qwen3 30B A3B986±161.9K3.4%5.1%163 tps1.0s41K$0.06$0.21
23126DeepSeek V3975±105.5K0.5%0.9%69 tps1.1s64K$0.59$1.49
2465Mistral Large 3974±201.2K5.5%2.1%51 tps1.0s256K$0.50$1.50
25129Command A959±86.2K1.2%2.2%42 tps0.8s256K$2.00$7.33
26165Pixtral Large941±189403.6%2.5%57 tps1.3s128K$1.50$4.50
27106DeepSeek V3.1 Terminus Thinking927±178404.5%5.9%27 tps1.8s131K$0.56$1.68
28161Mistral Small 3.1924±366152.4%7.4%13 tps2.6s32K$0.17$0.28
29148DeepSeek-R1904±161.7K2.8%0.8%133 tps0.6s64K$0.91$3.07
30161Llama 4 Maverick900±115.1K1.9%1.2%88 tps2.4s1M$0.23$0.83
31219NVIDIA Llama 3.3 Nemotron Super 49B v1883±179151.1%<0.1%13 tpsN/A131K$0.07$0.20
32201Gemma 3 27B IT879±215601.8%2.0%60 tps0.8s128K$0.17$0.29
33186GLM 4.6V Flash858±237502.6%3.7%64 tps2.1s128K$0.04$0.40
34177Mistral Small 3.1 24B Instruct839±226953.5%7.5%15 tps2.4s131K$0.06$0.18
35222Sky T1 32B Preview821±186251.6%7.8%73 tps0.6s16K$0.12$0.18
36292Arcee AI Spotlight788±151.1K1.3%<0.1%121 tps0.4s131K$0.18$0.18
37225Command R 7B787±266602.9%1.1%76 tps0.4s128K$0.04$0.15
38200NVIDIA Llama 3.1 Nemotron 70B783±171.2K2.4%<0.1%9 tps0.1s128K$0.33$0.39
39200K2 Think763±246050.8%<0.1%418 tps2.8sN/A$0$0
40246DeepSeek-R1 Distill Llama 70B755±199602.5%3.6%27 tps1.6s32K$0.73$0.95
View All (48 models)