Home/Speed ranking

Speed ranking

Output throughput in tok/s. For a 4,000-character answer: 17 seconds at 235 tok/s, 2 min 23 s at 28 tok/s.

#ModelValueVs. leaderComposite v5Output $Type
1 Claude Haiku 5.5Anthropic 235 #1 36.2 $0.50
2 gpt-oss-120b Open weightsOpenAI 220 94% — $0.20 Open weights
3 Gemma 4 31B Open weightsGoogle 200 85% — $0.40 Open weights
4 Qwen3.8 27B Open weightsAlibaba Qwen 180 77% — $0.40 Open weights
5 Gemini 3.8 FlashGoogle 160 68% — $3.75
6 DeepSeek V4.1 Flash Open weightsDeepSeek 148 63% 32.5 $1.20 Open weights
7 Muse WonderMeta 143 61% 43.3 $10.00
8 Claude Sonnet 5.5Anthropic 135 57% 54.2 $10.00
9 GPT-6 LunaOpenAI 130 55% — $0.50
10 MiniMax M3 Open weightsMiniMax 126 54% — $1.00 Open weights
11 GLM-5.3 Flash Open weightsZ.ai 120 51% — $0.80 Open weights
12 MiniMax M3.1 FlashMiniMax 118 50% 29.0 $0.60
13 Grok 4.7xAI 105 45% 35.0 $6.00
14 Claude Opus 5.5Anthropic 96 41% 58.4 $20.00
15 Step 5 PreviewStepFun 95 40% — $2.00
16 GPT-6.1 TerraOpenAI 94 40% 51.0 $5.00
17 Qwen3.8 MaxAlibaba Qwen 88 37% — $8.00
18 Gemini 4 CarbonGoogle 87 37% 59.4 $10.00
19 GPT-6 AstraOpenAI 81 34% 56.9 $50.00
20 MiMo-V2.6 Pro Open weightsXiaomi 81 34% 30.0 $1.40 Open weights
21 Gemini 4 ArgonGoogle 80 34% 50.4 $10.00
22 GPT-6.1 SolOpenAI 77 33% 54.5 $10.00
23 GLM-5.3 Open weightsZ.ai 77 33% 32.0 $2.20 Open weights
24 DeepSeek V4 Pro Open weightsDeepSeek 72 31% 23.0 $3.96 Open weights
25 Nemotron 3 Ultra Open weightsNVIDIA 70 30% — $1.60 Open weights
26 Claude Fable 5.1Anthropic 57 24% 49.7 $50.00
27 Kimi K3 Open weightsMoonshot AI 30 13% — $15.00 Open weights
28 Kimi K3.1 Pro Open weightsMoonshot AI 28 12% 46.0 $10.00 Open weights

← Back to the overall leaderboard