Price per 1M tokens using the Artificial Analysis 3:1 blended ratio.
Blended Price is a cost metric for comparing providers with different input and output token prices. Artificial Analysis calculates it with a 3:1 input-to-output token ratio so a single number can summarize typical API cost.
Test type: Provider price normalization. Lower values rank better.
371 models have this metric.
Current leader: Gemma 3n E4B Instruct
Project links
Prices come from the current Artificial Analysis pricing dataset, read from Supabase when configured.
Top models ranked by Blended $.
| Rank | Model | Creator | Value | Speed | Blended Price |
|---|---|---|---|---|---|
| #1 | Gemma 3n E4B Instruct | $0.025/M | 49.7 tok/s | $0.025/M | |
| #2 |
| Sarvam |
| $0.047/M |
| 200.7 tok/s |
| $0.047/M |
| #3 | Qwen3.5 4B (Non-reasoning) | Alibaba | $0.060/M | 23 tok/s | $0.060/M |
| #4 | Qwen3.5 4B (Reasoning) | Alibaba | $0.060/M | 22.8 tok/s | $0.060/M |
| #5 | Nova Micro | Amazon | $0.061/M | 249.8 tok/s | $0.061/M |
| #6 | Granite 4.1 8B | IBM | $0.063/M | 117.8 tok/s | $0.063/M |
| #7 | HyperNova 60B 2605 | Multiverse Computing | $0.065/M | 350.9 tok/s | $0.065/M |
| #8 | NVIDIA Nemotron Nano 9B V2 (Reasoning) | NVIDIA | $0.070/M | 85.9 tok/s | $0.070/M |
| #9 | Llama 3 Instruct 8B | Meta | $0.070/M | 79.8 tok/s | $0.070/M |
| #10 | Sarvam 105B (high) | Sarvam | $0.074/M | 115.9 tok/s | $0.074/M |
| #11 | Granite 3.3 8B (Non-reasoning) | IBM | $0.085/M | 414.9 tok/s | $0.085/M |
| #12 | NVIDIA Nemotron Nano 9B V2 (Non-reasoning) | NVIDIA | $0.086/M | 138.9 tok/s | $0.086/M |
| #13 | gpt-oss-20b (high) | OpenAI | $0.088/M | 224.5 tok/s | $0.088/M |
| #14 | NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | NVIDIA | $0.088/M | 99.8 tok/s | $0.088/M |
| #15 | NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | NVIDIA | $0.088/M | 111.7 tok/s | $0.088/M |
| #16 | Qwen2.5 Turbo | Alibaba | $0.088/M | 113.6 tok/s | $0.088/M |
| #17 | gpt-oss-20b (low) | OpenAI | $0.095/M | 260.9 tok/s | $0.095/M |
| #18 | Ministral 3 3B | Mistral | $0.100/M | 158.5 tok/s | $0.100/M |
| #19 | Llama 3.1 Instruct 8B | Meta | $0.100/M | 155.5 tok/s | $0.100/M |
| #20 | Llama 3.2 Instruct 1B | Meta | $0.100/M | 86.4 tok/s | $0.100/M |
| #21 | Llama 2 Chat 7B | Meta | $0.100/M | 95.6 tok/s | $0.100/M |
| #22 | Nova Lite | Amazon | $0.105/M | 147.8 tok/s | $0.105/M |
| #23 | Granite 4.0 H Small | IBM | $0.107/M | 439.5 tok/s | $0.107/M |
| #24 | Qwen3.5 9B (Reasoning) | Alibaba | $0.113/M | 67.6 tok/s | $0.113/M |
| #25 | Olmo 3 7B Instruct | Allen Institute for AI | $0.125/M | n/a | $0.125/M |
| #26 | Apertus 8B Instruct | Swiss AI Initiative | $0.125/M | n/a | $0.125/M |
| #27 | Nemotron 3 Nano Omni 30B A3B Reasoning | NVIDIA | $0.131/M | 320 tok/s | $0.131/M |
| #28 | GPT-5 nano (minimal) | OpenAI | $0.138/M | 168.7 tok/s | $0.138/M |
| #29 | GPT-5 nano (medium) | OpenAI | $0.138/M | 155.7 tok/s | $0.138/M |
| #30 | GPT-5 nano (high) | OpenAI | $0.138/M | 154.5 tok/s | $0.138/M |
| #31 | Gemma 4 12B (Non-reasoning) | $0.150/M | 121.2 tok/s | $0.150/M |
| #32 | Gemma 4 12B (Reasoning) | $0.150/M | 125.3 tok/s | $0.150/M |
| #33 | Ministral 3 8B | Mistral | $0.150/M | 89 tok/s | $0.150/M |
| #34 | Step 3.5 Flash 2603 | StepFun | $0.150/M | 256.9 tok/s | $0.150/M |
| #35 | Ling 2.6 Flash | InclusionAI | $0.150/M | 172.8 tok/s | $0.150/M |
| #36 | Llama 3.2 Instruct 3B | Meta | $0.150/M | 51.7 tok/s | $0.150/M |
| #37 | Mistral Small 3 | Mistral | $0.150/M | 145.3 tok/s | $0.150/M |
| #38 | Mistral Small 3.1 | Mistral | $0.150/M | 144.9 tok/s | $0.150/M |
| #39 | Mistral Small 3.2 | Mistral | $0.150/M | 144.9 tok/s | $0.150/M |
| #40 | Devstral Small (Jul '25) | Mistral | $0.150/M | n/a | $0.150/M |
| #41 | Solar Mini | Upstage | $0.150/M | 78.2 tok/s | $0.150/M |
| #42 | Step 3.5 Flash | StepFun | $0.150/M | 247.4 tok/s | $0.150/M |
| #43 | MiMo-V2-Flash (Reasoning) | Xiaomi | $0.150/M | n/a | $0.150/M |
| #44 | GLM-4.7-Flash (Reasoning) | Z AI | $0.153/M | 83.7 tok/s | $0.153/M |
| #45 | GLM-4.7-Flash (Non-reasoning) | Z AI | $0.153/M | 122.7 tok/s | $0.153/M |
| #46 | DeepSeek V4 Flash (Reasoning, Max Effort) | DeepSeek | $0.175/M | 101.1 tok/s | $0.175/M |
| #47 | DeepSeek V4 Flash (Reasoning, High Effort) | DeepSeek | $0.175/M | n/a | $0.175/M |
| #48 | DeepSeek V4 Flash (Non-reasoning) | DeepSeek | $0.175/M | 90.1 tok/s | $0.175/M |
| #49 | MiMo-V2.5 | Xiaomi | $0.175/M | 75.5 tok/s | $0.175/M |
| #50 | GPT-4.1 nano | OpenAI | $0.175/M | 148 tok/s | $0.175/M |
| #51 | Gemini 2.5 Flash-Lite (Reasoning) | $0.175/M | 237.8 tok/s | $0.175/M |
| #52 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) | $0.175/M | n/a | $0.175/M |
| #53 | Gemini 2.5 Flash-Lite (Non-reasoning) | $0.175/M | 182.3 tok/s | $0.175/M |
| #54 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) | $0.175/M | n/a | $0.175/M |
| #55 | Gemma 4 26B A4B (Non-reasoning) | $0.198/M | 70.8 tok/s | $0.198/M |
| #56 | Gemma 4 26B A4B (Reasoning) | $0.198/M | n/a | $0.198/M |
| #57 | Hermes 4 - Llama-3.1 70B (Non-reasoning) | Nous Research | $0.198/M | 96.2 tok/s | $0.198/M |
| #58 | Hermes 4 - Llama-3.1 70B (Reasoning) | Nous Research | $0.198/M | 92.1 tok/s | $0.198/M |
| #59 | Ministral 3 14B | Mistral | $0.200/M | 75.5 tok/s | $0.200/M |
| #60 | Hy3-preview (Reasoning) | Tencent | $0.200/M | 124.1 tok/s | $0.200/M |
| #61 | Hy3-preview (Non-reasoning) | Tencent | $0.200/M | 121.5 tok/s | $0.200/M |
| #62 | Gemma 4 31B (Non-reasoning) | $0.205/M | 43.7 tok/s | $0.205/M |
| #63 | Phi-4 | Microsoft | $0.219/M | 27 tok/s | $0.219/M |
| #64 | Ring-flash-2.0 | InclusionAI | $0.247/M | n/a | $0.247/M |
| #65 | Ling-flash-2.0 | InclusionAI | $0.247/M | 61.8 tok/s | $0.247/M |
| #66 | Mistral 7B Instruct | Mistral | $0.250/M | 91.5 tok/s | $0.250/M |
| #67 | Jamba 1.5 Mini | AI21 Labs | $0.250/M | n/a | $0.250/M |
| #68 | Jamba 1.6 Mini | AI21 Labs | $0.250/M | 179.6 tok/s | $0.250/M |
| #69 | gpt-oss-120b (high) | OpenAI | $0.262/M | 231.3 tok/s | $0.262/M |
| #70 | gpt-oss-120b (low) | OpenAI | $0.262/M | 296.3 tok/s | $0.262/M |
| #71 | Mistral Small 4 (Non-reasoning) | Mistral | $0.262/M | 142.1 tok/s | $0.262/M |
| #72 | Mistral Small 4 (Reasoning) | Mistral | $0.262/M | 166.7 tok/s | $0.262/M |
| #73 | GPT-4o mini | OpenAI | $0.262/M | 73.5 tok/s | $0.262/M |
| #74 | Gemini 2.0 Flash (Feb '25) | $0.262/M | n/a | $0.262/M |
| #75 | Qwen3.5 Omni Flash | Alibaba | $0.275/M | 227.8 tok/s | $0.275/M |
| #76 | Grok 4 Fast (Reasoning) | SpaceXAI | $0.275/M | n/a | $0.275/M |
| #77 | Grok 4 Fast (Non-reasoning) | SpaceXAI | $0.275/M | n/a | $0.275/M |
| #78 | Llama 4 Scout | Meta | $0.287/M | 89.6 tok/s | $0.287/M |
| #79 | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | NVIDIA | $0.300/M | 224.5 tok/s | $0.300/M |
| #80 | NVIDIA Nemotron Nano 12B v2 VL (Reasoning) | NVIDIA | $0.300/M | 282.5 tok/s | $0.300/M |