LLM API 제공업체 리더보드 - 500개 이상 AI 모델 엔드포인트 비교
OpenAI, Google, DeepSeek 등을 포함한 500개 이상의 AI 모델 엔드포인트에서 API 제공업체의 가격, 출력 속도, 지연 시간, 컨텍스트 창 등 주요 성능 지표를 비교합니다.
방법론을 비롯한 자세한 내용은 FAQ에서 확인하세요.
추가 분석 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
 Anthropic | Claude Opus 5.5 (max with fallback) | 1M | 독점 | 58 | $5.98 | 96 | 652.61 | 657.81 | -- | |
 Amazon Bedrock | Claude Opus 5.5 (max with fallback) | 1M | 독점 | 58 | $5.98 | 95 | 766.75 | 772.00 | -- | |
 Google | Claude Opus 5.5 (max with fallback) | 1M | 독점 | 58 | $11.63 | 96 | 676.59 | 681.81 | -- | |
 Microsoft Azure | Claude Opus 5.5 (max with fallback) | 1M | 독점 | 58 | $6.70 | 95 | 701.42 | 706.67 | -- | |
| Anthropic | Claude Sonnet 5.5 (max with fallback) | 1M | 독점 | 56 | $5.46 | 134 | 371.62 | 375.34 | -- | |
| Microsoft Azure | Claude Sonnet 5.5 (max with fallback) | 1M | 독점 | 56 | $4.96 | 129 | 402.38 | 406.27 | -- | |
| Amazon Bedrock | Claude Sonnet 5.5 (max with fallback) | 1M | 독점 | 56 | -- | 137 | 413.22 | 416.86 | -- | |
| Google | Claude Sonnet 5.5 (max with fallback) | 1M | 독점 | 56 | -- | 137 | 415.98 | 419.64 | -- | |
| Anthropic | Claude Opus 5.5 (xhigh with fallback) | 1M | 독점 | 56 | $3.46 | 82 | 109.62 | 115.71 | -- | |
 Kimi | Kimi K3.1 Pro | 1M | 독점 | 55 | $2.11 | 58 | 3.07 | 46.53 | 34.76 | |
| Kimi | Kimi K3.1 Pro High Speed | 1M | 독점 | 55 | -- | 112 | 2.44 | 24.74 | 17.84 | |
| Anthropic | Claude Opus 5.5 (high with fallback) | 1M | 독점 | 54 | $1.82 | 80 | 30.30 | 36.52 | -- | |
| Anthropic | Claude Fable 5.1 (max with fallback) | 1M | 독점 | 53 | $7.63 | 70 | 180.51 | 187.66 | -- | |
| Anthropic | Claude Fable 5.1 (xhigh with fallback) | 1M | 독점 | 53 | $5.98 | 63 | 74.48 | 82.45 | -- | |
 OpenAI | GPT-6 Astra (max) | 1M | 독점 | 53 | $3.26 | 66 | 278.25 | 285.78 | -- | |
| Microsoft Azure | GPT-6 Astra (max) | 524k | 독점 | 53 | $3.35 | 64 | 297.58 | 305.42 | -- | |
| Amazon Bedrock | GPT-6 Astra (max) | 524k | 독점 | 53 | $3.26 | 79 | 232.69 | 239.05 | -- | |
| OpenAI | GPT-6 Astra (max) (Fast) | 1M | 독점 | 53 | $6.52 | -- | -- | -- | -- | |
| OpenAI | GPT-6 Astra (max) (Ultrafast) | 1M | 독점 | 53 | $19.55 | -- | -- | -- | -- | |
| OpenAI | GPT-6 Astra (max) (Flex) | 1M | 독점 | 53 | $1.63 | 116 | 210.21 | 214.51 | -- | |
| Google | Gemini 4 Argon (high) AI Studio (1M output) | 1M | 독점 | 53 | $1.99 | -- | -- | -- | -- | |
| Google | Gemini 4 Argon (high) AI Studio | 1M | 독점 | 53 | $1.99 | 98 | 53.49 | 58.61 | -- | |
| Google | Gemini 4 Argon (high) (Vertex) | 1M | 독점 | 53 | $1.99 | 81 | 57.01 | 63.20 | -- | |
 Meta | Muse Wonder (max) | 1.05M | 독점 | 52 | $4.54 | 162 | 27.53 | 42.99 | 12.37 | |
| OpenAI | GPT-6 Astra (xhigh) | 1M | 독점 | 52 | $2.31 | 62 | 161.52 | 169.60 | -- | |
| Anthropic | Claude Sonnet 5.5 (xhigh with fallback) | 1M | 독점 | 52 | $2.01 | 108 | 27.33 | 31.96 | -- | |
| OpenAI | GPT-6.1 Sol (max) | 1M | 독점 | 52 | $0.72 | 59 | 236.01 | 244.51 | -- | |
| Microsoft Azure | GPT-6.1 Sol (max) | 1.05M | 독점 | 52 | $0.73 | 69 | 240.98 | 248.18 | -- | |
| OpenAI | GPT-6.1 Sol (max) (Fast) | 1M | 독점 | 52 | $1.45 | 108 | 160.47 | 165.09 | -- | |
| OpenAI | GPT-6.1 Sol (max) (Flex) | 1M | 독점 | 52 | $0.36 | 85 | 181.58 | 187.47 | -- | |
| OpenAI | GPT-6.1 Sol (max) (Ultrafast) | 1M | 독점 | 52 | $4.35 | 346 | 49.47 | 50.92 | -- | |
| Anthropic | Claude Opus 5.5 (medium with fallback) | 1M | 독점 | 51 | $1.34 | 83 | 15.65 | 21.68 | -- | |
| Anthropic | Claude Fable 5.1 (high with fallback) | 1M | 독점 | 51 | $3.91 | 58 | 20.04 | 28.67 | -- | |
| OpenAI | GPT-6.1 Sol (xhigh) | 1M | 독점 | 51 | $0.39 | 53 | 106.61 | 116.01 | -- | |
| OpenAI | GPT-6 Astra (high) | 1M | 독점 | 51 | $1.73 | 57 | 40.74 | 49.57 | -- | |
| OpenAI | GPT-6.1 Sol (high) | 1M | 독점 | 50 | $0.32 | 53 | 36.13 | 45.50 | -- | |
| OpenAI | GPT-6 Astra (medium) | 1M | 독점 | 50 | $1.54 | 53 | 5.13 | 14.54 | -- | |
| Anthropic | Claude Fable 5.1 (medium with fallback) | 1M | 독점 | 49 | $2.98 | 55 | 5.16 | 14.29 | -- | |
| OpenAI | GPT-6.1 Sol (medium) | 1M | 독점 | 48 | $0.21 | 51 | 4.02 | 13.77 | -- | |
| Anthropic | Claude Fable 5.1 (low with fallback) | 1M | 독점 | 47 | $2.37 | 57 | 1.73 | 10.44 | -- | |
| Anthropic | Claude Sonnet 5.5 (high with fallback) | 1M | 독점 | 47 | $0.88 | 106 | 10.46 | 15.18 | -- | |
 SpaceXAI | Grok 4.7 (xhigh) | 500k | 독점 | 46 | $3.74 | 93 | 40.25 | 45.65 | -- | |
| SpaceXAI | Grok 4.7 (high) | 500k | 독점 | 46 | $2.73 | 86 | 20.85 | 26.69 | -- | |
 Xiaomi | MiMo-V2.6-Pro | 1M | 오픈 | 46 | $0.13 | 57 | 3.11 | 47.12 | 35.20 | |
 DeepInfra | MiMo-V2.6-Pro | 1M | 오픈 | 46 | $0.96 | 40 | 2.11 | 64.51 | 49.92 | |
 PrimaLabs | MiMo-V2.6-Pro | 1.05M | 오픈 | 46 | $0.14 | 243 | 0.71 | 11.01 | 8.24 | |
 GMI | MiMo-V2.6-Pro (BF16) | 1M | 오픈 | 46 | $0.12 | 62 | 2.95 | 43.01 | 32.05 | |
| OpenAI | GPT-6 Astra (low) | 1M | 독점 | 46 | $0.82 | 53 | 2.32 | 11.77 | -- | |
| Microsoft Azure | GPT-6 Astra (low) | 524k | 독점 | 46 | $0.84 | 57 | 2.50 | 11.32 | -- | |
 Alibaba Cloud | Qwen3.8 Max (0902) | 984k | 독점 | 45 | $5.41 | 40 | 2.83 | 65.39 | 50.05 | |
 Baseten |  DeepSeek V4.1 Flash 1017 (max) | 1M | 오픈 | 45 | $1.01 | 301 | 0.35 | 8.66 | 6.65 | |
 Novita | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.24 | 161 | 1.43 | 16.94 | 12.41 | |
| Baseten | DeepSeek V4.1 Flash 1017 (max) (FAST) | 1.05M | 오픈 | 45 | $2.03 | 841 | 0.21 | 3.19 | 2.38 | |
 Systalyze | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.29 | 679 | 1.31 | 4.99 | 2.94 | |
| GMI | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.62 | 157 | 2.00 | 17.90 | 12.72 | |
| PrimaLabs | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.23 | 287 | 0.65 | 9.37 | 6.97 | |
 LithosAI | DeepSeek V4.1 Flash 1017 (max) (ULTRA CHAT) | 1.05M | 오픈 | 45 | $0.37 | 1,267 | 0.52 | 2.49 | 1.58 | |
| LithosAI | DeepSeek V4.1 Flash 1017 (max) (ULTRA) | 1.05M | 오픈 | 45 | $0.37 | 469 | 1.17 | 6.50 | 4.27 | |
| LithosAI | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.16 | 300 | 0.94 | 9.28 | 6.67 | |
| LithosAI | DeepSeek V4.1 Flash 1017 (max) (FAST) | 1.05M | 오픈 | 45 | $0.26 | 429 | 1.45 | 7.28 | 4.67 | |
| DeepSeek | DeepSeek V4.1 Flash 1017 (max) | 1M | 오픈 | 45 | $0.26 | 169 | 1.18 | 15.97 | 11.83 | |
| DeepInfra | DeepSeek V4.1 Flash 1017 (max) | 1M | 오픈 | 45 | $0.61 | 66 | 0.78 | 38.80 | 30.41 | |
 Fireworks | DeepSeek V4.1 Flash 1017 (max) | 1M | 오픈 | 45 | $0.24 | 206 | 1.12 | 13.28 | 9.73 | |
 Parasail | DeepSeek V4.1 Flash 1017 (max) | 1M | 오픈 | 45 | $0.24 | 155 | 0.67 | 16.77 | 12.87 | |
 Together AI | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.32 | 373 | 0.30 | 7.01 | 5.37 | |
 SiliconFlow | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.83 | 135 | 1.99 | 20.57 | 14.86 | |
 Makora | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $1.05 | 246 | 0.68 | 10.86 | 8.14 | |
 Nebius | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.26 | 437 | 1.33 | 7.05 | 4.58 | |
 Inco | DeepSeek V4.1 Flash 1017 (max) (FAST) | 1.05M | 오픈 | 45 | $2.14 | 1,077 | 1.62 | 3.94 | 1.86 | |
 DigitalOcean | DeepSeek V4.1 Flash 1017 (max) | 1M | 오픈 | 45 | $1.68 | 324 | 0.52 | 8.23 | 6.17 | |
 Wafer | DeepSeek V4.1 Flash 1017 (max) | 1.05M | 오픈 | 45 | $0.28 | 189 | 0.56 | 13.75 | 10.56 | |
 Modal |  GLM-5.3 (max) | 1M | 오픈 | 45 | $1.97 | 100 | 0.74 | 25.67 | 19.94 | |
| Zai | GLM-5.3 (max) | 1M | 오픈 | 45 | $2.01 | 98 | 2.39 | 27.86 | 20.37 | |
| Together AI | GLM-5.3 (max) | 1.05M | 오픈 | 45 | $1.97 | 160 | 0.67 | 16.27 | 12.48 | |
| Parasail | GLM-5.3 (max) | 1M | 오픈 | 45 | $1.97 | 201 | 1.06 | 13.50 | 9.95 | |
| DeepInfra | GLM-5.3 (max) | 1.05M | 오픈 | 45 | $0.97 | 88 | 0.97 | 29.30 | 22.66 | |
 Modular | GLM-5.3 (max) | 1.05M | 오픈 | 45 | $2.96 | 173 | 0.67 | 15.12 | 11.56 | |
| Inco | GLM-5.3 (max) (FAST) | 1M | 오픈 | 45 | $4.08 | 687 | 0.75 | 4.39 | 2.91 | |
| Novita | GLM-5.3 (max) | 1M | 오픈 | 45 | $2.00 | 93 | 2.16 | 28.97 | 21.45 | |
| Baseten | GLM-5.3 (max) | 1.05M | 오픈 | 45 | $1.43 | 116 | 3.43 | 24.92 | 17.19 | |
| DigitalOcean | GLM-5.3 (max) | 1.05M | 오픈 | 45 | $1.97 | 57 | 0.91 | 44.96 | 35.24 | |
| Makora | GLM-5.3 (max) | 1.05M | 오픈 | 45 | $0.92 | 210 | 0.93 | 12.81 | 9.51 | |
 Databricks | GLM-5.3 (max) | 1M | 오픈 | 45 | $2.73 | 185 | 0.85 | 14.36 | 10.81 | |
| Wafer | GLM-5.3 (max) | 1M | 오픈 | 45 | $2.04 | 174 | 0.76 | 15.12 | 11.49 | |
| Baseten | GLM-5.3 (max) (FAST) | 1M | 오픈 | 45 | $3.75 | 251 | 1.00 | 10.98 | 7.98 | |
| LithosAI | GLM-5.3 (max) (FP4) | 1.05M | 오픈 | 45 | $2.35 | 475 | 0.90 | 6.16 | 4.21 | |
| LithosAI | GLM-5.3 (max) (FP4 ULTRA CHAT) | 1.05M | 오픈 | 45 | $5.14 | 746 | 144.57 | 147.92 | 2.68 | |
| Inco | GLM-5.3 (max) | 164k | 오픈 | 45 | $2.04 | 85 | 2.44 | 31.75 | 23.44 | |
 Bitdeer AI | GLM-5.3 (max) | 1M | 오픈 | 45 | $4.99 | 157 | 1.60 | 17.50 | 12.72 | |
 Crusoe | GLM-5.3 (max) (FP4) | 786k | 오픈 | 45 | $3.36 | 229 | 0.73 | 11.62 | 8.72 | |
 Supernova | GLM-5.3 (max) (FP4) | 1M | 오픈 | 45 | $1.19 | 235 | 1.88 | 12.50 | 8.50 | |
 FriendliAI | GLM-5.3 (max) | 1M | 오픈 | 45 | $1.81 | 148 | 0.82 | 17.65 | 13.47 | |
| Supernova | GLM-5.3 (max) (FP4, FAST) | 1M | 오픈 | 45 | $5.29 | 607 | 1.47 | 5.59 | 3.29 | |
| Nebius | GLM-5.3 (max) (FP4) | 1M | 오픈 | 45 | $1.97 | 209 | 1.47 | 13.42 | 9.56 | |
| GMI | GLM-5.3 (max) | 1M | 오픈 | 45 | $1.42 | 117 | 2.71 | 24.00 | 17.03 | |
 Multiverse Computing | GLM-5.3 (max) | 1M | 오픈 | 45 | $0.80 | 198 | 1.44 | 14.09 | 10.12 | |
 Scaleway | GLM-5.3 (max) | 262k | 오픈 | 45 | $12.94 | 251 | 1.30 | 11.25 | 7.96 | |
 InclusionAI | Ling 3.1 Flash 1019 | 1M | 독점 | 44 | $1.49 | -- | -- | -- | -- | |
 StepFun | Step 5 Preview | 1M | 오픈 | 44 | $1.03 | 229 | 1.15 | 12.08 | 8.74 | |
| Anthropic | Claude Haiku 5.5 (max) | 1M | 독점 | 43 | $0.21 | 234 | 322.16 | 324.30 | -- | |
| DigitalOcean |  MiniMax-M3.1-Flash | 1.05M | 독점 | 43 | $0.33 | 375 | 0.69 | 7.36 | 5.34 | |
| GMI | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.37 | 111 | 1.64 | 24.09 | 17.96 | |
| Nebius | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.24 | 104 | 1.74 | 25.82 | 19.27 | |
| Inco | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.32 | 99 | 1.22 | 26.59 | 20.30 | |
| Parasail | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.20 | 271 | 0.79 | 10.02 | 7.38 | |
| Together AI | MiniMax-M3.1-Flash | 1.05M | 독점 | 43 | $0.29 | 238 | 0.59 | 11.09 | 8.40 | |
| Wafer | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.19 | 175 | 1.30 | 15.61 | 11.45 | |
| Fireworks | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.22 | 178 | 0.82 | 14.88 | 11.25 | |
| Inco | MiniMax-M3.1-Flash | 262k | 독점 | 43 | $0.63 | 778 | 1.46 | 4.67 | 2.57 | |
| SiliconFlow | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.19 | 110 | 3.39 | 26.10 | 18.17 | |
| Modular | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.19 | 226 | 0.44 | 11.51 | 8.85 | |
| Novita | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.20 | 127 | 1.88 | 21.61 | 15.79 | |
| MiniMax | MiniMax-M3.1-Flash | 1M | 독점 | 43 | $0.22 | 113 | 1.35 | 23.52 | 17.74 | |
| Anthropic | Claude Opus 5.5 (low with fallback) | 1M | 독점 | 42 | $0.55 | 82 | 6.23 | 12.34 | -- | |
| Amazon Bedrock | Claude Opus 5.5 (low with fallback) | 1M | 독점 | 42 | $0.55 | 79 | 5.99 | 12.32 | -- | |
| Google | Claude Opus 5.5 (low with fallback) | 1M | 독점 | 42 | $0.68 | 83 | 6.13 | 12.14 | -- | |
| Microsoft Azure | Claude Opus 5.5 (low with fallback) | 1M | 독점 | 42 | $0.59 | 83 | 4.34 | 10.37 | -- | |
| SpaceXAI | Grok 4.7 (low) | 500k | 독점 | 42 | $1.25 | 78 | 6.25 | 12.66 | -- | |
| OpenAI | GPT-6.1 Sol (low) | 1M | 독점 | 42 | $0.13 | 52 | 2.05 | 11.60 | -- | |
| OpenAI | GPT-5.6 Terra (max) | 1M | 독점 | 42 | $1.40 | 130 | 102.74 | 106.58 | -- | |
| Amazon Bedrock | GPT-5.6 Terra (max) | 1M | 독점 | 42 | $1.41 | 146 | 111.32 | 114.75 | -- | |
| Modal | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.88 | 121 | 0.73 | 21.39 | 16.52 | |
| Zai | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.28 | 70 | 2.84 | 38.43 | 28.48 | |
| Baseten | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.60 | 130 | 0.41 | 19.64 | 15.38 | |
| DeepInfra | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.34 | 64 | 0.80 | 39.72 | 31.13 | |
| Novita | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.29 | 79 | 3.22 | 34.82 | 25.28 | |
| Fireworks | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.37 | 167 | 3.16 | 18.16 | 12.00 | |
| Databricks | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.28 | 143 | 0.92 | 18.44 | 14.02 | |
| SiliconFlow | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.50 | 65 | 1.98 | 40.66 | 30.94 | |
| GMI | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.31 | 55 | 3.95 | 49.06 | 36.09 | |
| Bitdeer AI | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.24 | 254 | 1.43 | 11.26 | 7.87 | |
| Wafer | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.45 | 145 | 0.70 | 17.94 | 13.80 | |
| Together AI | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.47 | 173 | 0.72 | 15.18 | 11.57 | |
| Parasail | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.67 | 125 | 1.25 | 21.20 | 15.96 | |
| FriendliAI | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.51 | 246 | 0.60 | 10.76 | 8.13 | |
| Nebius | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $1.19 | 108 | 1.57 | 24.67 | 18.48 | |
| DigitalOcean | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.42 | 267 | 0.66 | 10.01 | 7.48 | |
| Makora | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.54 | 146 | 1.00 | 18.07 | 13.65 | |
 CoreWeave | GLM-5.3-Flash | 1M | 오픈 | 42 | $0.65 | 267 | 1.38 | 10.76 | 7.50 | |
| Crusoe | GLM-5.3-Flash (FP4) | 1.05M | 오픈 | 42 | $0.81 | 156 | 1.07 | 17.09 | 12.82 | |
| Inco | GLM-5.3-Flash (FAST) | 1M | 오픈 | 42 | $0.56 | 1,338 | 1.90 | 3.77 | 1.49 | |
| LithosAI | GLM-5.3-Flash (FP4) | 1.05M | 오픈 | 42 | $0.29 | 372 | 0.83 | 7.55 | 5.37 | |
| LithosAI | GLM-5.3-Flash (FP4, ULTRA CHAT) | 1.05M | 오픈 | 42 | $0.56 | 1,469 | 0.42 | 2.12 | 1.36 | |
| PrimaLabs | GLM-5.3-Flash | 1.05M | 오픈 | 42 | $0.44 | 264 | 0.69 | 10.18 | 7.59 | |
| LithosAI | GLM-5.3-Flash (FP4, ULTRA) | 1.05M | 오픈 | 42 | $0.57 | 509 | 0.87 | 5.79 | 3.93 | |
| Anthropic | Claude Haiku 5.5 (xhigh) | 1M | 독점 | 41 | $0.12 | 186 | 59.51 | 62.19 | -- | |
| DeepInfra | Ling 3.1 Flash | 262k | 오픈 | 41 | $1.01 | 11 | 0.91 | 225.34 | 179.54 | |
| InclusionAI | Ling 3.1 Flash | 1M | 오픈 | 41 | $1.01 | 213 | 1.82 | 13.56 | 9.39 | |
| Google | Gemini 3.8 Flash (high) AI Studio | 1M | 독점 | 41 | $1.24 | 137 | 27.26 | 30.91 | -- | |
| Anthropic | Claude Sonnet 5.5 (medium with fallback) | 1M | 독점 | 41 | $0.48 | 106 | 1.35 | 6.07 | -- | |
| Fireworks | Qwen3.8 2.4T A95B | 262k | 오픈 | 40 | $1.87 | 158 | 0.88 | 16.72 | 12.67 | |
| Together AI | Qwen3.8 2.4T A95B | 262k | 오픈 | 40 | $2.09 | 151 | 1.06 | 17.67 | 13.29 | |
| DigitalOcean | Qwen3.8 2.4T A95B | 262k | 오픈 | 40 | $1.80 | 62 | 0.84 | 41.08 | 32.18 | |
| Modal | Qwen3.8 2.4T A95B | 1M | 오픈 | 40 | $1.96 | 203 | 0.74 | 13.07 | 9.86 | |
| DeepInfra | Qwen3.8 2.4T A95B | 262k | 오픈 | 40 | $2.89 | 88 | 1.00 | 29.51 | 22.81 | |
| Alibaba Cloud | Qwen3.8 2.4T A95B | 262k | 오픈 | 40 | $2.16 | 41 | 2.94 | 64.18 | 48.99 | |
| Alibaba Cloud | Qwen3.8-Flash-Next | 1M | 오픈 | 40 | $0.37 | 52 | 2.12 | 50.52 | 38.72 | |
| Google | Gemini 3.8 Flash (medium) AI Studio | 1M | 독점 | 40 | $0.93 | -- | -- | -- | -- | |
| Fireworks | Muse Spark Open (xhigh) | 1.05M | 오픈 | 39 | $0.84 | 85 | 0.76 | 30.14 | 23.50 | |
| Baseten | Muse Spark Open (xhigh) | 1.05M | 오픈 | 39 | $0.84 | 242 | 0.25 | 10.59 | 8.27 | |
| Nebius | Muse Spark Open (xhigh) (FP4) | 1.05M | 오픈 | 39 | $0.70 | 132 | 1.35 | 20.28 | 15.14 | |
| Together AI | Muse Spark Open (xhigh) | 1.05M | 오픈 | 39 | $0.84 | 203 | 0.67 | 12.98 | 9.85 | |
| Makora | Muse Spark Open (xhigh) (NVFP4) | 1.05M | 오픈 | 39 | $0.72 | 381 | 0.89 | 7.44 | 5.24 | |
 Mistral | Mistral Large 4 Preview | 524k | 독점 | 38 | $1.13 | 109 | 1.78 | 24.70 | 18.33 | |
| OpenAI | GPT-6 Luna (max) | 1M | 독점 | 38 | $0.07 | 140 | 78.46 | 82.03 | -- | |
| Microsoft Azure | GPT-6 Luna (max) | 524k | 독점 | 38 | $0.07 | 190 | 60.22 | 62.85 | -- | |
| OpenAI | GPT-6 Luna (max) (Flex) | 1M | 독점 | 38 | $0.03 | 45 | 209.10 | 220.13 | -- | |
| OpenAI | GPT-5.6 Terra (xhigh) | 1M | 독점 | 38 | $0.63 | 118 | 15.08 | 19.30 | -- | |
| Microsoft Azure | GPT-5.6 Terra (xhigh) | 524k | 독점 | 38 | $0.71 | 134 | 15.15 | 18.87 | -- | |
| Xiaomi | MiMo-V2.6-Flash | 1M | 오픈 | 38 | $0.06 | 56 | 3.79 | 48.63 | 35.87 | |
| DeepInfra | MiMo-V2.6-Flash | 1M | 오픈 | 38 | $0.42 | 32 | 8.47 | 87.52 | 63.24 | |
| GMI | MiMo-V2.6-Flash (BF16) | 524k | 오픈 | 38 | $0.06 | 54 | 3.66 | 49.75 | 36.87 | |
| Makora | MiMo-V2.6-Flash | 1.05M | 오픈 | 38 | $0.22 | 141 | 0.96 | 18.74 | 14.22 | |
| Anthropic | Claude Haiku 5.5 (high) | 1M | 독점 | 38 | $0.08 | 181 | 25.00 | 27.77 | -- | |
| GMI | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $1.76 | 70 | 2.83 | 38.66 | 28.66 | |
| Crusoe | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $1.65 | 189 | 1.47 | 14.69 | 10.57 | |
| DeepSeek | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $0.67 | 105 | 1.75 | 25.65 | 19.12 | |
| SiliconFlow | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $1.45 | 62 | 1.56 | 41.84 | 32.23 | |
| Baseten | DeepSeek V4 Pro 0813 (max) | 1M | 오픈 | 36 | $5.51 | 133 | 1.65 | 20.45 | 15.04 | |
| Novita | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $0.70 | 83 | 2.09 | 32.18 | 24.07 | |
| DigitalOcean | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $1.61 | 57 | 0.91 | 44.67 | 35.01 | |
| DeepInfra | DeepSeek V4 Pro 0813 (max) | 1M | 오픈 | 36 | $1.00 | 102 | 0.97 | 25.44 | 19.57 | |
| Systalyze | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $0.51 | 303 | 1.47 | 9.71 | 6.60 | |
| Nebius | DeepSeek V4 Pro 0813 (max) | 1.05M | 오픈 | 36 | $3.63 | 188 | 1.60 | 14.90 | 10.64 | |
| Google | Claude Sonnet 5.5 (low with fallback) | 1M | 독점 | 36 | -- | 108 | 1.19 | 5.83 | -- | |
| Anthropic | Claude Sonnet 5.5 (low with fallback) | 1M | 독점 | 36 | $0.35 | 105 | 0.98 | 5.73 | -- | |
| Amazon Bedrock | Claude Sonnet 5.5 (low with fallback) | 1M | 독점 | 36 | -- | 107 | 1.49 | 6.18 | -- | |
| Microsoft Azure | Claude Sonnet 5.5 (low with fallback) | 1M | 독점 | 36 | $0.33 | 102 | 0.86 | 5.78 | -- | |
 Agnes AI | Agnes 2.5 Pro Beta | 1M | 독점 | 35* | -- | 114 | 1.39 | 23.29 | 17.52 | |
| DeepInfra | DeepSeek V4 Flash Vision (max) | 1M | 독점 | 35 | $0.51 | 194 | 0.75 | 13.65 | 10.31 | |
| GMI | DeepSeek V4 Flash Vision (max) | 1M | 독점 | 35 | $0.28 | 171 | 2.33 | 16.96 | 11.71 | |
| SiliconFlow | DeepSeek V4 Flash Vision (max) | 1M | 독점 | 35 | $2.51 | 170 | 1.43 | 16.12 | 11.75 | |
| OpenAI | GPT-6 Luna (xhigh) | 1M | 독점 | 35 | $0.04 | 127 | 19.84 | 23.77 | -- | |
| Anthropic | Claude Haiku 5.5 (medium) | 1M | 독점 | 34 | $0.05 | 179 | 9.73 | 12.53 | -- | |
| Zai | GLM-5.3 (low) | 1M | 오픈 | 34 | $0.85 | 91 | 3.02 | 30.49 | 21.98 | |
| Supernova | GLM-5.3 (low) (FP4) | 1M | 오픈 | 34 | $0.50 | 257 | 1.91 | 11.63 | 7.78 | |
| Supernova | GLM-5.3 (low) (FP4, FAST) | 1M | 오픈 | 34 | $2.24 | 603 | 1.54 | 5.69 | 3.32 | |
| Bitdeer AI | GLM-5.3 (low) | 1M | 오픈 | 34 | $2.11 | 160 | 1.63 | 17.27 | 12.51 | |
| DigitalOcean | GLM-5.3 (low) | 1.05M | 오픈 | 34 | $0.84 | 57 | 0.84 | 44.98 | 35.31 | |
| Scaleway | GLM-5.3 (low) | 262k | 오픈 | 34 | $5.46 | 268 | 1.28 | 10.59 | 7.45 | |
| Inco | GLM-5.3 (low) | 164k | 오픈 | 34 | $0.87 | 92 | 2.91 | 29.95 | 21.63 | |
| Inco | GLM-5.3 (low) (FAST) | 1M | 오픈 | 34 | $1.73 | 703 | 0.76 | 4.31 | 2.85 | |
| OpenAI | GPT-5.6 Terra (high) | 1M | 독점 | 34 | $0.34 | 119 | 2.23 | 6.43 | -- | |
| Microsoft Azure | GPT-5.6 Terra (high) | 524k | 독점 | 34 | $0.37 | 126 | 2.74 | 6.71 | -- | |
| DeepInfra | Qwen3.8 27B (xhigh) | 262k | 오픈 | 34 | $0.52 | 30 | 1.15 | 85.44 | 67.43 | |
| CoreWeave | Qwen3.8 27B (xhigh) (FP8) | 262k | 오픈 | 34 | $0.92 | 44 | 1.93 | 58.44 | 45.21 | |
| Alibaba Cloud | Qwen3.8 27B (xhigh) | 1M | 오픈 | 34 | $1.01 | 45 | 3.88 | 59.12 | 44.20 | |
| Parasail | Qwen3.8 27B (xhigh) (FP8) | 262k | 오픈 | 34 | $0.60 | 75 | 1.20 | 34.45 | 26.60 | |
| Multiverse Computing | Qwen3.8 27B (xhigh) | 262k | 오픈 | 34 | -- | 185 | 0.88 | 14.38 | 10.80 | |
| Wafer | Qwen3.8 27B (xhigh) | 262k | 오픈 | 34 | $0.61 | 92 | 1.14 | 28.38 | 21.79 | |
| Crusoe | Qwen3.8 27B (xhigh) | 262k | 오픈 | 34 | $0.84 | 212 | 0.85 | 12.62 | 9.41 | |
| Nebius | Qwen3.8 27B (xhigh) (FP4) | 262k | 오픈 | 34 | $0.70 | 399 | 1.38 | 7.65 | 5.01 | |
| Modular | Qwen3.8 27B (xhigh) (NVFP4) | 262k | 오픈 | 34 | $0.00 | 214 | 0.57 | 12.26 | 9.35 | |
| Google | Gemini 3.8 Flash (low) AI Studio | 1M | 독점 | 33 | -- | -- | -- | -- | -- | |
| OpenAI | GPT-6 Luna (high) | 1M | 독점 | 33 | $0.03 | 120 | 10.51 | 14.68 | -- | |
| Microsoft Azure | GPT-6 Luna (high) | 524k | 독점 | 33 | $0.03 | -- | -- | -- | -- | |
| OpenAI | GPT-5.3 Codex (xhigh) | 400k | 독점 | 33* | -- | 99 | 54.85 | 59.88 | -- | |
 Institute of Foundation Models | K2 Horizon 375B A23B | 524k | 오픈 | 31 | $0.00 | 879 | 2.16 | 5.01 | 2.28 | |
| Baseten |  Beam 1.0 (NVFP4) | 1.05M | 오픈 | 30 | $0.40 | 516 | 0.45 | 5.30 | 3.88 | |
| CoreWeave | Beam 1.0 (NVFP4) | 1.05M | 오픈 | 30 | $0.92 | 364 | 1.00 | 7.86 | 5.49 | |
| Together AI | Beam 1.0 (NVFP4) | 1.05M | 오픈 | 30 | $0.21 | 337 | 0.54 | 7.97 | 5.94 | |
| DigitalOcean | Beam 1.0 | 1.05M | 오픈 | 30 | $0.12 | 233 | 0.62 | 11.35 | 8.59 | |
| Nebius | Beam 1.0 (FP8) | 1.05M | 오픈 | 30 | $0.30 | 372 | 1.25 | 7.97 | 5.38 | |
| OpenAI | GPT-5.6 Terra (medium) | 1M | 독점 | 30 | $0.18 | 114 | 1.49 | 5.85 | -- | |
| OpenAI | GPT-6 Luna (medium) | 1M | 독점 | 30 | $0.02 | 92 | 2.79 | 8.25 | -- | |
| Google | Gemini 3.1 Pro Preview (Vertex) | 1M | 독점 | 30 | $1.31 | 97 | 25.64 | 30.81 | -- | |
| Google | Gemini 3.1 Pro Preview (AI Studio) | 1M | 독점 | 30 | $1.30 | 132 | 20.63 | 24.41 | -- | |
| Anthropic | Claude Haiku 5.5 (low) | 1M | 독점 | 29 | $0.02 | 173 | 7.52 | 10.41 | -- | |
| Parasail | MiniMax-M3 (MXFP8) | 1M | 오픈 | 29 | $0.46 | 162 | 1.02 | 16.48 | 12.37 | |
| CoreWeave | MiniMax-M3 (NVFP4) | 262k | 오픈 | 29 | -- | 185 | 1.15 | 14.65 | 10.80 | |
| Together AI | MiniMax-M3 | 1M | 오픈 | 29 | $0.58 | 88 | 0.88 | 29.45 | 22.85 | |
| Novita | MiniMax-M3 | 1M | 오픈 | 29 | $0.46 | 100 | 1.49 | 26.51 | 20.02 | |
| Modular | MiniMax-M3 | 1M | 오픈 | 29 | $0.46 | 148 | 1.33 | 18.21 | 13.50 | |
| Nebius | MiniMax-M3 (FP8) | 1.05M | 오픈 | 29 | $0.51 | 344 | 1.45 | 8.72 | 5.81 | |
| SiliconFlow | MiniMax-M3 (FP8) | 1M | 오픈 | 29 | $0.45 | 88 | 1.52 | 29.86 | 22.67 | |
| MiniMax | MiniMax-M3 | 512k | 오픈 | 29 | $0.51 | 103 | 1.30 | 25.47 | 19.33 | |
| DeepInfra | MiniMax-M3 | 524k | 오픈 | 29 | $0.44 | 21 | 2.83 | 119.36 | 93.22 | |
| GMI | MiniMax-M3 | 1.05M | 오픈 | 29 | -- | 103 | 1.20 | 25.47 | 19.41 | |
 SambaNova | MiniMax-M3 | 1.05M | 오픈 | 29 | $3.83 | 199 | 2.38 | 14.92 | 10.03 | |
| Fireworks | MiniMax-M3 | 524k | 오픈 | 29 | $0.50 | 175 | 0.89 | 15.18 | 11.43 | |
| Alibaba Cloud | Qwen3.8 27B (medium) | 1M | 오픈 | 28 | $1.13 | 48 | 4.01 | 56.04 | 41.62 | |
| OpenAI | GPT-5.6 Terra (low) | 1M | 독점 | 27 | $0.14 | 116 | 1.35 | 5.67 | -- | |
| Multiverse Computing | Quasar 438B (max) | 1M | 독점 | 27 | -- | 159 | 1.25 | 16.94 | 12.56 | |
 Upstage | Solar Pro 4 | 512k | 독점 | 27 | $0.71 | -- | -- | -- | -- | |
| PrimaLabs |  Nemotron 3.5 Super VL (NVFP4) | 1.05M | 오픈 | 26 | $0.18 | 603 | 0.86 | 5.01 | 3.32 | |
| Baseten | Nemotron 3.5 Super VL | 1M | 오픈 | 26 | $0.11 | 541 | 0.29 | 4.90 | 3.69 | |
| Nebius | Nemotron 3.5 Super VL | 1.05M | 오픈 | 26 | $0.44 | 414 | 1.29 | 7.33 | 4.83 | |
| CoreWeave | Nemotron 3.5 Super VL (NVFP4) | 1M | 오픈 | 26 | $0.24 | 517 | 1.46 | 6.30 | 3.87 | |
| DigitalOcean | Nemotron 3.5 Super VL | 131k | 오픈 | 26 | $0.35 | 181 | 0.66 | 14.49 | 11.06 | |
| Crusoe | Nemotron 3.5 Super VL | 262k | 오픈 | 26 | $0.16 | 334 | 0.58 | 8.06 | 5.99 | |
| Together AI | Nemotron 3.5 Super VL (NVFP4) | 262k | 오픈 | 26 | $0.26 | 433 | 0.58 | 6.35 | 4.62 | |
 Apodex | Apodex 1.1 | 256k | 독점 | 26 | $0.46 | -- | -- | -- | -- | |
| Alibaba Cloud | Qwen3.8 27B (low) | 1M | 오픈 | 26 | $1.05 | 49 | 3.92 | 55.26 | 41.08 | |
| OpenAI | GPT-5.5 Instant (June 2026) | 400k | 독점 | 26 | $0.69 | 152 | 0.90 | 17.30 | 13.12 | |
| Nebius | Kimi K2.7 Code (FP4) | 256k | 오픈 | 26 | $0.46 | 215 | 1.36 | 14.05 | 10.37 | |
| GMI | Kimi K2.7 Code (FP8) | 65.5k | 오픈 | 26 | $1.37 | 65 | 5.40 | 47.70 | 34.55 | |
| CoreWeave | Kimi K2.7 Code | 262k | 오픈 | 26 | $0.83 | 101 | 1.35 | 28.43 | 22.11 | |
| Novita | Kimi K2.7 Code | 262k | 오픈 | 26 | $0.55 | 72 | 3.86 | 41.82 | 31.01 | |
| DeepInfra | Kimi K2.7 Code | 262k | 오픈 | 26 | $0.39 | 63 | 1.14 | 44.80 | 35.66 | |
| Kimi | Kimi K2.7 Code | 256k | 오픈 | 26 | $0.54 | 64 | 2.79 | 45.47 | 34.86 | |
| DeepInfra |  Inkling Small | 524k | 오픈 | 26 | -- | 325 | 0.71 | 8.40 | 6.15 | |
| Thinking Machines | Inkling Small | 256k | 오픈 | 26 | $0.09 | 223 | 1.58 | 12.80 | 8.97 | |
| SiliconFlow |  Hy3 (FP8) | 256k | 오픈 | 25 | $0.07 | 89 | 3.00 | 31.12 | 22.49 | |
| Novita | Hy3 | 262k | 오픈 | 25 | $0.07 | 85 | 2.63 | 32.06 | 23.54 | |
| DeepInfra | Hy3 (FP8) | 262k | 오픈 | 25 | $0.07 | 51 | 1.73 | 50.38 | 38.92 | |
| GMI | Hy3 | 262k | 오픈 | 25 | $0.07 | 96 | 2.86 | 28.93 | 20.86 | |
| Alibaba Cloud | Qwen3.7 Plus | 1M | 독점 | 25 | $0.22 | 56 | 2.30 | 47.01 | 35.77 | |
| DeepInfra | Inkling (xhigh) (FP8) | 131k | 오픈 | 25 | -- | 344 | 0.73 | 8.00 | 5.82 | |
| Together AI | Inkling (xhigh) | 524k | 오픈 | 25 | $0.91 | 105 | 0.67 | 24.58 | 19.13 | |
| Modal | Inkling (xhigh) (NVFP4) | 1.05M | 오픈 | 25 | $0.86 | 199 | 0.97 | 13.53 | 10.05 | |
| Thinking Machines | Inkling (xhigh) | 262k | 오픈 | 25 | -- | 195 | 1.79 | 14.58 | 10.23 | |
| Databricks | Inkling (xhigh) | 1M | 오픈 | 25 | $0.61 | 220 | 1.34 | 12.72 | 9.10 | |
| Fireworks | Inkling (xhigh) | 1.05M | 오픈 | 25 | $0.63 | 183 | 0.69 | 14.33 | 10.91 | |
| Upstage | Solar Mini 4 | 524k | 독점 | 24 | $0.37 | 112 | 1.85 | 24.23 | 17.91 | |
| CoreWeave | Nemotron 3 Ultra | 262k | 오픈 | 23 | $0.51 | 354 | 1.00 | 8.83 | 6.42 | |
| DeepInfra | Nemotron 3 Ultra | 262k | 오픈 | 23 | $0.39 | 65 | 1.17 | 44.04 | 35.15 | |
| DeepInfra | Nemotron 3 Ultra BF16 | 262k | 오픈 | 23 | $0.99 | 96 | 1.23 | 30.18 | 23.74 | |
| Nebius | Nemotron 3 Ultra | 256k | 오픈 | 23 | $0.59 | 341 | 2.17 | 10.31 | 6.67 | |
| Fireworks | Nemotron 3 Ultra | 262k | 오픈 | 23 | $0.32 | 141 | 0.68 | 20.29 | 16.08 | |
| Google | Gemini 3.5 Flash-Lite AI Studio | 1M | 독점 | 22 | $0.19 | 414 | 7.85 | 9.05 | -- | |
| OpenAI | GPT-6 Luna (low) | 1M | 독점 | 22 | $0.0045 | 116 | 2.45 | 6.75 | -- | |
| DeepInfra | Qwen3.5 397B A17B (non-reasoning) (FP8) | 262k | 오픈 | 21* | -- | 50 | 1.18 | 11.25 | -- | |
| Alibaba Cloud | Qwen3.5 397B A17B (non-reasoning) | 262k | 오픈 | 21* | -- | 60 | 2.15 | 10.54 | -- | |
| Nebius | Qwen3.5 397B A17B (non-reasoning) (Base, FP4) | 262k | 오픈 | 21* | -- | 145 | 1.63 | 5.08 | -- | |
| DigitalOcean | Qwen3.5 397B A17B (non-reasoning) | 131k | 오픈 | 21* | -- | 21 | 1.39 | 25.08 | -- | |
| Novita | Qwen3.5 397B A17B (non-reasoning) | 262k | 오픈 | 21* | -- | 59 | 2.22 | 10.62 | -- | |
| OpenAI | GPT-5.6 Terra (non-reasoning) | 1M | 독점 | 21 | $0.14 | 122 | 0.75 | 4.85 | -- | |
| Amazon Bedrock | GPT-5.6 Terra (non-reasoning) | 1M | 독점 | 21 | $0.14 | 125 | 0.52 | 4.51 | -- | |
| Alibaba Cloud | Qwen3.5 Omni Plus | 256k | 독점 | 20* | -- | 122 | 2.04 | 6.16 | -- | |
| Systalyze | DeepSeek V4 Pro 0813 (non-reasoning) | 1.05M | 오픈 | 20 | $0.26 | 289 | 1.41 | 3.15 | -- | |
| DeepSeek | DeepSeek V4 Pro 0813 (non-reasoning) | 1M | 오픈 | 20 | $0.36 | 100 | 1.76 | 6.74 | -- | |
| Nebius | DeepSeek V4 Pro 0813 (non-reasoning) | 1.05M | 오픈 | 20 | $1.85 | 195 | 1.61 | 4.18 | -- | |
| DigitalOcean | DeepSeek V4 Pro 0813 (non-reasoning) | 1.05M | 오픈 | 20 | $0.43 | 58 | 0.94 | 9.63 | -- | |
| Microsoft Azure | o3 | 200k | 독점 | 20* | -- | 111 | 24.67 | 29.16 | -- | |
| OpenAI | o3 | 200k | 독점 | 20* | -- | 152 | 7.15 | 10.44 | -- | |
| Alibaba Cloud | Qwen3.8 27B (non-reasoning) | 1M | 오픈 | 20 | $2.49 | 50 | 3.97 | 13.99 | -- | |
| Multiverse Computing | Qwen3.8 27B (non-reasoning) | 262k | 오픈 | 20 | -- | 208 | 0.89 | 3.29 | -- | |
| Wafer | Qwen3.8 27B (non-reasoning) | 262k | 오픈 | 20 | $1.34 | 99 | 1.22 | 6.26 | -- | |
| Crusoe | Qwen3.8 27B (non-reasoning) | 262k | 오픈 | 20 | $2.18 | 229 | 0.89 | 3.07 | -- | |
| Modular | Qwen3.8 27B (non-reasoning) (NVFP4) | 262k | 오픈 | 20 | $0.00 | 258 | 0.60 | 2.54 | -- | |
| Nebius | Qwen3.8 27B (non-reasoning) (FP4) | 262k | 오픈 | 20 | $1.54 | 429 | 1.39 | 2.55 | -- | |
| SiliconFlow |  LongCat 2.0 (FP8) | 262k | 오픈 | 19 | $0.15 | 76 | 2.57 | 35.67 | 26.48 | |
| OpenAI | GPT-6 Luna (non-reasoning) | 1M | 독점 | 18 | $0.01 | 120 | 0.87 | 5.03 | -- | |
| Microsoft Azure | GPT-6 Luna (non-reasoning) | 524k | 독점 | 18 | $0.01 | 170 | 1.07 | 4.01 | -- | |
| Alibaba Cloud | Qwen3.5 397B A17B | 262k | 오픈 | 18 | $0.47 | 59 | 2.15 | 64.73 | 54.09 | |
| Parasail | Qwen3.5 397B A17B | 262k | 오픈 | 18 | $0.29 | 85 | 1.19 | 44.44 | 37.38 | |
| DeepInfra | Qwen3.5 397B A17B (FP8) | 262k | 오픈 | 18 | -- | 49 | 1.12 | 75.92 | 64.66 | |
| Nebius | Qwen3.5 397B A17B (Base, FP4) | 262k | 오픈 | 18 | $0.17 | 142 | 1.62 | 27.50 | 22.37 | |
| Novita | Qwen3.5 397B A17B | 262k | 오픈 | 18 | $0.47 | 59 | 2.39 | 64.75 | 53.90 | |
| DigitalOcean | Qwen3.5 397B A17B | 131k | 오픈 | 18 | $0.17 | 24 | 1.33 | 155.78 | 133.50 | |
| Alibaba Cloud | Qwen3.6 35B A3B | 229k | 오픈 | 18 | $0.48 | 131 | 2.05 | 46.92 | 41.07 | |
| SiliconFlow | Qwen3.6 35B A3B (FP8) | 262k | 오픈 | 18 | $0.32 | 119 | 1.21 | 50.84 | 45.42 | |
| DeepInfra | Qwen3.6 35B A3B (FP8) | 262k | 오픈 | 18 | $0.14 | 115 | 0.78 | 51.92 | 46.80 | |
| CoreWeave | Qwen3.6 35B A3B (FP8) | 262k | 오픈 | 18 | $0.31 | 175 | 1.09 | 34.86 | 30.90 | |
| GMI | Qwen3.6 35B A3B FP8 | 262k | 오픈 | 18 | $0.31 | 110 | 1.84 | 55.36 | 48.98 | |
| Parasail | Qwen3.6 35B A3B | 262k | 오픈 | 18 | $0.12 | 116 | 0.82 | 51.46 | 46.34 | |
| Novita | Qwen3.6 35B A3B | 205k | 오픈 | 18 | $0.31 | 135 | 1.94 | 45.63 | 39.98 | |
| Scaleway | Qwen3.6 35B A3B | 262k | 오픈 | 18 | $0.37 | 144 | 1.11 | 42.16 | 37.57 | |
| DeepInfra | Qwen3.5 122B A10B (non-reasoning) (FP4) | 262k | 오픈 | 18* | -- | 51 | 1.16 | 10.97 | -- | |
| Alibaba Cloud | Qwen3.5 122B A10B (non-reasoning) | 262k | 오픈 | 18* | -- | 143 | 2.30 | 5.78 | -- | |
| Together AI | Muse Glimmer (high) | 131k | 오픈 | 17 | $0.08 | 89 | 0.76 | 28.88 | 22.50 | |
| DeepInfra | Muse Glimmer (high) | 131k | 오픈 | 17 | $0.05 | 88 | 1.03 | 29.50 | 22.77 | |
| Systalyze | Muse Glimmer (high) | 131k | 오픈 | 17 | $0.03 | 173 | 1.49 | 15.98 | 11.59 | |
| Novita | Gemma 4 26B A4B | 262k | 오픈 | 17* | -- | 90 | 1.27 | 29.19 | 22.33 | |
| Google | Gemma 4 26B A4B AI Studio | 262k | 오픈 | 17* | -- | 47 | 0.95 | 54.64 | 42.96 | |
| Systalyze | Gemma 4 26B A4B | 131k | 오픈 | 17* | -- | 481 | 0.94 | 6.14 | 4.16 | |
| CoreWeave | Gemma 4 26B A4B | 262k | 오픈 | 17* | -- | 69 | 0.81 | 36.91 | 28.88 | |
| DeepInfra | Gemma 4 26B A4B (FP8) | 262k | 오픈 | 17* | -- | 82 | 0.64 | 31.20 | 24.44 | |
| Parasail | Gemma 4 26B A4B | 256k | 오픈 | 17* | -- | 92 | 1.20 | 28.42 | 21.78 | |
| Makora | Gemma 4 26B A4B | 1M | 오픈 | 17* | -- | 202 | 0.60 | 13.00 | 9.91 | |
| InclusionAI | Ring-2.6-1T | 262k | 오픈 | 17 | $0.29 | 124 | 3.70 | 23.91 | 16.17 | |
| SiliconFlow | Qwen3.5 122B A10B (FP8) | 262k | 오픈 | 16 | $0.21 | 72 | 1.70 | 36.23 | 27.62 | |
| DeepInfra | Qwen3.5 122B A10B (FP4) | 262k | 오픈 | 16 | $0.24 | 47 | 1.14 | 54.53 | 42.71 | |
| Alibaba Cloud | Qwen3.5 122B A10B | 262k | 오픈 | 16 | $0.32 | 128 | 2.28 | 21.75 | 15.57 | |
| Novita | Qwen3.5 122B A10B | 262k | 오픈 | 16 | $0.32 | 129 | 2.37 | 21.80 | 15.55 | |
| Scaleway | Qwen3.6 35B A3B (non-reasoning) | 262k | 오픈 | 15* | -- | 139 | 1.10 | 4.68 | -- | |
| CoreWeave | Qwen3.6 35B A3B (non-reasoning) (FP8) | 262k | 오픈 | 15* | -- | 161 | 1.10 | 4.21 | -- | |
| DeepInfra | Qwen3.6 35B A3B (non-reasoning) (FP8) | 262k | 오픈 | 15* | -- | 96 | 0.76 | 5.96 | -- | |
| Novita | Qwen3.6 35B A3B (non-reasoning) | 205k | 오픈 | 15* | -- | 149 | 1.90 | 5.25 | -- | |
| GMI | Qwen3.6 35B A3B (non-reasoning) FP8 | 262k | 오픈 | 15* | -- | 91 | 1.85 | 7.34 | -- | |
| Parasail | Qwen3.6 35B A3B (non-reasoning) (FP8) | 262k | 오픈 | 15* | -- | 130 | 0.93 | 4.77 | -- | |
| Alibaba Cloud | Qwen3.6 35B A3B (non-reasoning) | 262k | 오픈 | 15* | -- | 147 | 2.04 | 5.44 | -- | |
| CoreWeave | Gemma 4 31B | 262k | 오픈 | 15 | $0.07 | 64 | 2.11 | 37.12 | 27.18 | |
| Modular | Gemma 4 31B (NVFP4) | 262k | 오픈 | 15 | $0.10 | 296 | 0.43 | 7.98 | 5.87 | |
| Novita | Gemma 4 31B | 262k | 오픈 | 15 | $0.10 | 27 | 2.41 | 85.47 | 64.49 | |
| Google | Gemma 4 31B (AI Studio) | 262k | 오픈 | 15 | $0.00 | 35 | 1.08 | 64.27 | 49.06 | |
| Parasail | Gemma 4 31B | 262k | 오픈 | 15 | $0.07 | 81 | 1.07 | 28.78 | 21.52 | |
| DeepInfra | Gemma 4 31B | 262k | 오픈 | 15 | $0.15 | 23 | 1.51 | 99.92 | 76.40 | |
| Crusoe | Gemma 4 31B | 262k | 오픈 | 15 | $0.10 | 63 | 0.93 | 36.40 | 27.54 | |
| SiliconFlow | Gemma 4 31B (FP8) | 262k | 오픈 | 15 | $0.54 | 9 | 3.30 | 260.72 | 199.86 | |
| SambaNova | Gemma 4 31B | 131k | 오픈 | 15 | $0.28 | 488 | 3.26 | 7.84 | 3.56 | |
| FriendliAI | Gemma 4 31B | 256k | 오픈 | 15 | $0.10 | 139 | 0.76 | 16.88 | 12.52 | |
| DigitalOcean | Gemma 4 31B | 262k | 오픈 | 15 | $0.05 | 42 | 2.07 | 55.59 | 41.55 | |
| Mistral | Mistral Medium 3.5 | 262k | 오픈 | 14 | $0.50 | 162 | 2.35 | 17.73 | 12.31 | |
| SiliconFlow | Gemma 4 12B | 262k | 오픈 | 14* | -- | 114 | 2.24 | 24.13 | 17.51 | |
| Amazon Bedrock | Nova 2.0 Pro Preview (medium) | 256k | 독점 | 14* | -- | 127 | 12.50 | 32.24 | 15.79 | |
| Modular | Gemma 4 31B (non-reasoning) (NVFP4) | 262k | 오픈 | 14* | -- | 290 | 0.39 | 2.11 | -- | |
| Parasail | Gemma 4 31B (non-reasoning) | 262k | 오픈 | 14* | -- | 79 | 1.02 | 7.39 | -- | |
| FriendliAI | Gemma 4 31B (non-reasoning) | 262k | 오픈 | 14* | -- | 154 | 0.72 | 3.96 | -- | |
| CoreWeave | Gemma 4 31B (non-reasoning) | 262k | 오픈 | 14* | -- | 61 | 2.13 | 10.27 | -- | |
| SiliconFlow | Gemma 4 31B (non-reasoning) (FP8) | 262k | 오픈 | 14* | -- | 8 | 3.40 | 63.22 | -- | |
| SambaNova | Gemma 4 31B (non-reasoning) | 131k | 오픈 | 14* | -- | 205 | 3.65 | 6.10 | -- | |
| Crusoe | Gemma 4 31B (non-reasoning) | 262k | 오픈 | 14* | -- | 55 | 0.86 | 9.95 | -- | |
| Novita | Gemma 4 31B (non-reasoning) | 262k | 오픈 | 14* | -- | 24 | 2.19 | 23.13 | -- | |
| DigitalOcean | Gemma 4 31B (non-reasoning) | 262k | 오픈 | 14* | -- | 42 | 2.06 | 14.11 | -- | |
| DeepInfra | Gemma 4 31B (non-reasoning) (FP8) | 262k | 오픈 | 14* | -- | 26 | 1.38 | 20.61 | -- | |
| Amazon Bedrock | Nova 2.0 Lite (high) | 1M | 독점 | 13* | -- | 186 | 16.58 | 30.00 | 10.74 | |
| Together AI | Qwen3.5 9B (non-reasoning) (FP8) | 262k | 오픈 | 13* | -- | 80 | 0.90 | 7.17 | -- | |
 Cohere | Command A+ | 200k | 오픈 | 13 | $0.00 | 166 | 0.50 | 15.52 | 12.01 | |
| Parasail | Gemma 4 26B A4B (non-reasoning) | 262k | 오픈 | 13* | -- | 93 | 1.19 | 6.55 | -- | |
| Systalyze | Gemma 4 26B A4B (non-reasoning) | 131k | 오픈 | 13* | -- | 530 | 0.95 | 1.89 | -- | |
| Scaleway | Gemma 4 26B A4B (non-reasoning) | 262k | 오픈 | 13* | -- | 179 | 0.83 | 3.61 | -- | |
| SiliconFlow | Gemma 4 26B A4B (non-reasoning) (FP8) | 262k | 오픈 | 13* | -- | 17 | 2.23 | 31.29 | -- | |
| CoreWeave | Gemma 4 26B A4B (non-reasoning) | 262k | 오픈 | 13* | -- | 80 | 0.78 | 7.04 | -- | |
| DeepInfra | Gemma 4 26B A4B (non-reasoning) (FP8) | 262k | 오픈 | 13* | -- | 73 | 0.72 | 7.53 | -- | |
| Novita | Gemma 4 26B A4B (non-reasoning) | 262k | 오픈 | 13* | -- | 96 | 1.20 | 6.42 | -- | |
| Makora | Gemma 4 26B A4B (non-reasoning) | 1M | 오픈 | 13* | -- | 183 | 0.63 | 3.37 | -- | |
| DeepInfra | Qwen3.5 4B (FP8) | 262k | 오픈 | 13* | -- | 19 | 0.92 | 129.42 | 102.80 | |
| DeepInfra | Nemotron 3.5 Lightning (NVFP4) | 262k | 오픈 | 13 | $0.09 | 166 | 0.64 | 15.73 | 12.07 | |
| Nebius | Nemotron 3.5 Lightning (BF16) | 1.05M | 오픈 | 13 | $0.09 | 269 | 1.02 | 10.30 | 7.42 | |
| CoreWeave | Nemotron 3.5 Lightning (BF16) | 262k | 오픈 | 13 | $0.07 | 385 | 0.64 | 7.14 | 5.20 | |
| Fireworks | Nemotron 3.5 Lightning (BF16) | 262k | 오픈 | 13 | $0.02 | 388 | 0.46 | 6.91 | 5.15 | |
| PrimaLabs | Nemotron 3.5 Lightning (NVFP4) | 1.05M | 오픈 | 13 | $0.07 | 722 | 0.58 | 4.05 | 2.77 | |
| Crusoe | Nemotron 3.5 Lightning (BF16) | 262k | 오픈 | 13 | $0.08 | 284 | 0.48 | 9.28 | 7.04 | |
| Amazon Bedrock | Nova 2.0 Pro Preview (low) | 256k | 독점 | 13* | -- | 120 | 10.60 | 31.43 | 16.67 | |
| DeepInfra |  Granite 4.2 30B | 131k | 오픈 | 13 | $0.07 | 47 | 0.82 | 53.46 | 42.11 | |
| Amazon Bedrock | Nova 2.0 Lite (medium) | 1M | 독점 | 12* | -- | 180 | 12.97 | 26.88 | 11.13 | |
| Alibaba Cloud | Qwen3.5 Omni Flash | 256k | 독점 | 12* | -- | 232 | 1.79 | 3.95 | -- | |
 Inception | Mercury 2.5 | 256k | 독점 | 12 | $0.12 | 714 | 3.56 | 4.26 | -- | |
| Amazon Bedrock | Nova 2.0 Lite (low) | 1M | 독점 | 12* | -- | 173 | 9.17 | 23.65 | 11.58 | |
 Groq | gpt-oss-120b (high) | 131k | 오픈 | 12 | -- | 473 | 0.64 | 5.93 | 4.23 | |
| Scaleway | gpt-oss-120b (high) | 128k | 오픈 | 12 | $0.12 | 123 | 1.27 | 21.53 | 16.21 | |
| CoreWeave | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.02 | 82 | 1.35 | 31.89 | 24.43 | |
| Novita | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.04 | 239 | 0.97 | 11.43 | 8.37 | |
| DeepInfra | gpt-oss-120b (high) (Turbo) | 131k | 오픈 | 12 | $0.11 | 55 | 0.65 | 46.33 | 36.54 | |
| Microsoft Azure | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.11 | 379 | 0.74 | 7.33 | 5.27 | |
| DeepInfra | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.03 | 56 | 0.62 | 45.03 | 35.53 | |
 Cerebras | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.23 | 1,647 | 0.39 | 1.90 | 1.21 | |
| Google | gpt-oss-120b (high) Vertex | 131k | 오픈 | 12 | $0.06 | 173 | 0.70 | 15.18 | 11.58 | |
| Together AI | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.11 | 117 | 0.54 | 21.91 | 17.10 | |
| Crusoe | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.04 | 302 | 0.43 | 8.71 | 6.62 | |
| Baseten | gpt-oss-120b (high) | 128k | 오픈 | 12 | $0.07 | 208 | 0.18 | 12.22 | 9.64 | |
| SambaNova | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.15 | 701 | 1.13 | 4.69 | 2.85 | |
| Parasail | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.06 | 311 | 0.73 | 8.78 | 6.44 | |
| Nebius | gpt-oss-120b (high) (Base) | 128k | 오픈 | 12 | $0.11 | 358 | 0.96 | 7.94 | 5.59 | |
| Amazon Bedrock | gpt-oss-120b (high) | 131k | 오픈 | 12 | $0.11 | 99 | 1.00 | 26.29 | 20.23 | |
| Mistral | Mistral Small 4 | 256k | 오픈 | 11 | $0.02 | 178 | 0.86 | 14.88 | 11.22 | |
| SiliconFlow | Qwen3.5 9B (FP8) | 262k | 오픈 | 11 | $0.16 | 23 | 2.46 | 112.65 | 88.16 | |
| Together AI | Qwen3.5 9B (FP8) | 262k | 오픈 | 11 | $0.27 | 81 | 0.75 | 31.62 | 24.69 | |
| CoreWeave | Granite 4.2 8B | 131k | 오픈 | 11 | $0.05 | 87 | 0.78 | 29.47 | 22.95 | |
| DeepInfra | Granite 4.2 8B | 131k | 오픈 | 11 | $0.02 | 54 | 0.78 | 46.99 | 36.97 | |
| InclusionAI | Ling 3.0 Tiny | 262k | 오픈 | 11 | $0.00 | 58 | 2.61 | 45.89 | 34.62 | |
 Arcee AI | Trinity Large Thinking | 262k | 오픈 | 11 | $0.12 | 353 | 1.26 | 8.33 | 5.66 | |
| DeepInfra | Qwen3.5 4B (non-reasoning) FP8 | 262k | 오픈 | 11* | -- | 14 | 0.91 | 37.29 | -- | |
| Crusoe | Nemotron 3 Nano Omni 30B A3B | 262k | 오픈 | 10* | -- | 237 | 0.35 | 10.91 | 8.45 | |
| DigitalOcean | Nemotron 3 Nano Omni 30B A3B | 65.5k | 오픈 | 10* | -- | 230 | 0.97 | 11.83 | 8.69 | |
| Nebius | gpt-oss-120b (low) Base | 128k | 오픈 | 10* | -- | 366 | 0.91 | 7.75 | 5.47 | |
| SambaNova | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 742 | 1.10 | 4.47 | 2.70 | |
| Google | gpt-oss-120b (low) Vertex | 131k | 오픈 | 10* | -- | 116 | 0.59 | 22.18 | 17.28 | |
| Groq | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 469 | 0.64 | 5.97 | 4.27 | |
| Crusoe | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 328 | 0.45 | 8.07 | 6.10 | |
| DeepInfra | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 53 | 0.61 | 47.85 | 37.79 | |
| CoreWeave | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 74 | 1.35 | 35.15 | 27.04 | |
| Cerebras | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 1,694 | 0.38 | 1.86 | 1.18 | |
| Microsoft Azure | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 416 | 0.72 | 6.74 | 4.81 | |
| Baseten | gpt-oss-120b (low) | 128k | 오픈 | 10* | -- | 197 | 0.18 | 12.87 | 10.16 | |
| Amazon Bedrock | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 102 | 1.04 | 25.61 | 19.66 | |
| Parasail | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 316 | 0.76 | 8.67 | 6.33 | |
| Together AI | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 119 | 0.47 | 21.47 | 16.80 | |
| Novita | gpt-oss-120b (low) | 131k | 오픈 | 10* | -- | 218 | 0.93 | 12.41 | 9.18 | |
| Amazon Bedrock | Nova 2.0 Pro Preview (non-reasoning) | 256k | 독점 | 10* | -- | 115 | 1.03 | 5.37 | -- | |
| Groq | gpt-oss-20b (low) | 131k | 오픈 | 10* | -- | 891 | 0.78 | 3.59 | 2.24 | |
| CoreWeave | gpt-oss-20b (low) | 131k | 오픈 | 10* | -- | 170 | 0.61 | 15.32 | 11.77 | |
| Amazon Bedrock | gpt-oss-20b (low) | 131k | 오픈 | 10* | -- | 112 | 10.77 | 33.08 | 17.85 | |
| Cohere | North Mini Code | 256k | 오픈 | 10 | $0.00 | 66 | 0.45 | 38.22 | 30.22 | |
| Databricks | Qwen3 Next 80B A3B | 262k | 오픈 | 10* | -- | 47 | 0.78 | 11.33 | -- | |
| DeepInfra | Qwen3 Next 80B A3B | 262k | 오픈 | 10* | -- | 139 | 0.69 | 4.29 | -- | |
| Parasail | Qwen3 Next 80B A3B | 262k | 오픈 | 10* | -- | 172 | 0.79 | 3.70 | -- | |
| SiliconFlow | Gemma 4 12B (non-reasoning) | 262k | 오픈 | 9* | -- | 118 | 2.27 | 6.52 | -- | |
| Novita | Llama 4 Maverick (FP8) | 1.05M | 오픈 | 9 | $0.04 | 46 | 1.00 | 11.87 | -- | |
| Microsoft Azure | Llama 4 Maverick (FP8) | 128k | 오픈 | 9 | $0.04 | 379 | 1.32 | 2.63 | -- | |
| Amazon Bedrock | Llama 4 Maverick | 128k | 오픈 | 9 | $0.04 | -- | -- | -- | -- | |
| DeepInfra | Llama 4 Maverick (FP8) | 1.05M | 오픈 | 9 | $0.03 | 42 | 0.90 | 12.73 | -- | |
| DigitalOcean | Llama 4 Maverick | 128k | 오픈 | 9 | $0.01 | 17 | 0.99 | 30.50 | -- | |
| Parasail | Llama 4 Maverick (FP8) | 1.05M | 오픈 | 9 | $0.03 | 90 | 0.97 | 6.53 | -- | |
| Amazon Bedrock | Qwen3 Coder Next | 128k | 오픈 | 9 | $0.78 | 118 | 1.07 | 5.30 | -- | |
| Parasail | Qwen3 Coder Next (FP8) | 262k | 오픈 | 9 | $0.12 | 91 | 0.80 | 6.31 | -- | |
| DeepInfra | Granite 4.2 3B | 131k | 오픈 | 9 | $0.01 | 51 | 0.73 | 49.42 | 38.95 | |
| Mistral | Mistral Small 4 (non-reasoning) | 262k | 오픈 | 9* | -- | 173 | 0.92 | 3.81 | -- | |
| CoreWeave | gpt-oss-20b (high) | 131k | 오픈 | 9 | $0.01 | 163 | 0.63 | 15.97 | 12.28 | |
| Amazon Bedrock | gpt-oss-20b (high) | 131k | 오픈 | 9 | $0.01 | 175 | 31.65 | 45.95 | 11.44 | |
| DeepInfra | gpt-oss-20b (high) | 131k | 오픈 | 9 | $0.01 | 123 | 0.58 | 20.92 | 16.28 | |
| Groq | gpt-oss-20b (high) | 131k | 오픈 | 9 | $0.02 | 875 | 0.73 | 3.58 | 2.29 | |
| DeepInfra | Gemma 4 E4B | 131k | 오픈 | 9 | $0.01 | 45 | 0.91 | 56.02 | 44.09 | |
| Nebius | Nemotron 3 Nano | 262k | 오픈 | 9 | $0.02 | 159 | 1.01 | 16.73 | 12.57 | |
| Novita | Nemotron 3 Nano (FP4) | 262k | 오픈 | 9 | $0.02 | 277 | 0.99 | 10.01 | 7.22 | |
| DeepInfra | Nemotron 3 Nano | 262k | 오픈 | 9 | $0.02 | 149 | 0.61 | 17.44 | 13.46 | |
| Crusoe | Nemotron 3 Nano (FP8) | 262k | 오픈 | 9 | $0.01 | 287 | 0.37 | 9.09 | 6.97 | |
 Sarvam | Sarvam 105B (high) | 128k | 오픈 | 9* | -- | 189 | 1.59 | 14.79 | 10.56 | |
| Amazon Bedrock | Nova 2.0 Lite (non-reasoning) | 1M | 독점 | 9* | -- | 183 | 1.03 | 3.76 | -- | |
| Amazon Bedrock | Magistral Small 1.2 | 128k | 오픈 | 9* | -- | 77 | 1.49 | 33.99 | 26.00 | |
| Alibaba Cloud | Qwen3 Omni 30B A3B | 65.5k | 오픈 | 8* | -- | 103 | 1.89 | 26.25 | 19.49 | |
| Groq | Llama 3.3 70B | 131k | 오픈 | 8* | -- | 320 | 0.85 | 2.42 | -- | |
| Novita | Llama 3.3 70B | 131k | 오픈 | 8* | -- | 46 | 1.73 | 12.66 | -- | |
| SambaNova | Llama 3.3 70B | 128k | 오픈 | 8* | -- | 305 | 1.92 | 3.56 | -- | |
| Parasail | Llama 3.3 70B (FP8) | 131k | 오픈 | 8* | -- | 78 | 2.33 | 8.76 | -- | |
| CoreWeave | Llama 3.3 70B | 128k | 오픈 | 8* | -- | 93 | 0.88 | 6.23 | -- | |
| Together AI | Llama 3.3 70B Turbo | 131k | 오픈 | 8* | -- | 67 | 1.87 | 9.30 | -- | |
| Microsoft Azure | Llama 3.3 70B | 128k | 오픈 | 8* | -- | 129 | 2.14 | 6.03 | -- | |
| DeepInfra | Llama 3.3 70B (Turbo, FP8) | 131k | 오픈 | 8* | -- | 18 | 2.14 | 29.75 | -- | |
| Amazon Bedrock | Llama 3.3 70B | 128k | 오픈 | 8* | -- | -- | -- | -- | -- | |
| Scaleway | Llama 3.3 70B | 100k | 오픈 | 8* | -- | 82 | 1.31 | 7.38 | -- | |
| Nebius |  Hermes 4 405B (FP8) | 128k | 오픈 | 7* | -- | 43 | 2.30 | 60.58 | 46.63 | |
| DeepInfra | Gemma 4 E4B (non-reasoning) | 131k | 오픈 | 7* | -- | 45 | 0.79 | 11.87 | -- | |
| DeepInfra | NVIDIA Nemotron Nano 9B V2 | 131k | 오픈 | 7* | -- | 122 | 0.64 | 21.08 | 16.35 | |
| Nebius | Hermes 4 405B (non-reasoning) (FP8) | 128k | 오픈 | 7* | -- | 43 | 2.28 | 13.82 | -- | |
| Cohere | Command A | 288k | 오픈 | 7* | -- | 33 | 1.81 | 16.84 | -- | |
| Microsoft Azure | Command A | 256k | 오픈 | 7* | -- | 41 | 3.04 | 15.16 | -- | |
| Novita | Nemotron 3 Nano (non-reasoning) (FP4) | 262k | 오픈 | 7* | -- | 277 | 0.98 | 2.79 | -- | |
| DeepInfra | Nemotron 3 Nano (non-reasoning) | 262k | 오픈 | 7* | -- | 163 | 0.75 | 3.82 | -- | |
| Crusoe | Nemotron 3 Nano (non-reasoning) (FP8) | 262k | 오픈 | 7* | -- | 285 | 0.38 | 2.14 | -- | |
| DeepInfra | NVIDIA Nemotron Nano 9B V2 (non-reasoning) | 131k | 오픈 | 7* | -- | 122 | 0.58 | 4.68 | -- | |
| Amazon Bedrock | NVIDIA Nemotron Nano 9B V2 (non-reasoning) | 131k | 오픈 | 7* | -- | 170 | 1.07 | 4.02 | -- | |
| Microsoft Azure | Llama 4 Scout | 128k | 오픈 | 6 | $0.12 | 133 | 0.83 | 4.60 | -- | |
| Google | Llama 4 Scout Vertex | 1.31M | 오픈 | 6 | $0.16 | 158 | 0.72 | 3.88 | -- | |
| DeepInfra | Llama 4 Scout | 131k | 오픈 | 6 | $0.06 | 38 | 0.62 | 13.79 | -- | |
| Amazon Bedrock | Llama 4 Scout | 128k | 오픈 | 6 | $0.11 | -- | -- | -- | -- | |
| Novita | Llama 4 Scout | 131k | 오픈 | 6 | $0.11 | 38 | 1.00 | 14.02 | -- | |
 Celeris | Celeris-1 | 131k | 독점 | 6 | $0.05 | 1,619 | 0.56 | 0.87 | -- | |
| Microsoft Azure | Phi-4 Mini | 128k | 오픈 | 6* | -- | 46 | 0.85 | 11.83 | -- | |
| Mistral | Ministral 3 14B | 262k | 오픈 | 6 | $0.02 | 77 | 0.97 | 7.46 | -- | |
| Amazon Bedrock | Ministral 3 14B | 256k | 오픈 | 6 | $0.14 | 189 | 1.05 | 3.69 | -- | |
| DigitalOcean | Ministral 3 14B | 262k | 오픈 | 6 | $0.03 | 37 | 1.08 | 14.66 | -- | |
| Alibaba Cloud | Qwen3 Omni 30B A3B | 65.5k | 오픈 | 6* | -- | 96 | 1.82 | 7.01 | -- | |
| Microsoft Azure | Phi-4 | 16.4k | 오픈 | 6* | -- | 39 | 2.53 | 15.36 | -- | |
| DeepInfra | Phi-4 | 16.4k | 오픈 | 6* | -- | 95 | 1.10 | 6.39 | -- | |
| Amazon Bedrock | Nova Micro | 130k | 독점 | 6* | -- | 290 | 0.80 | 2.53 | -- | |
| Amazon Bedrock | NVIDIA Nemotron Nano 12B v2 VL (non-reasoning) | 128k | 오픈 | 6* | -- | 202 | 1.05 | 3.52 | -- | |
| Microsoft Azure | Phi-4 Multimodal | 4.1k | 오픈 | 6* | -- | -- | -- | -- | -- | |
 Reka AI | Reka Flash 3 | 54k | 오픈 | 6* | -- | 88 | 1.40 | 29.78 | 22.71 | |
| Amazon Bedrock | Ministral 3 8B | 256k | 오픈 | 5 | $0.07 | 250 | 0.97 | 2.97 | -- | |
| Mistral | Ministral 3 8B | 262k | 오픈 | 5 | $0.01 | 119 | 0.85 | 5.05 | -- | |
| DeepInfra | Llama 3.2 11B (Vision) | 131k | 오픈 | 5* | -- | 26 | 1.34 | 20.30 | -- | |
| Amazon Bedrock | Ministral 3 3B | 256k | 오픈 | 5 | $0.04 | 428 | 0.87 | 2.04 | -- | |
| Mistral | Ministral 3 3B | 131k | 오픈 | 5 | $0.01 | 209 | 0.66 | 3.06 | -- | |
| Cohere | Tiny Aya Global | 8.19k | 오픈 | 5* | -- | -- | -- | -- | -- | |
 Public AI |  Apertus 8B Instruct | 64k | 오픈 | 5* | -- | 136 | 3.52 | 7.21 | -- | |
| OpenAI | GPT-6 Sol (Daybreak Blue, max) | 1.05M | 독점 | -- | -- | -- | -- | -- | -- | |