# What 1M LLM tokens costs

Price per million tokens for 357 models, cheapest first. "1M tokens" here is 75% input and 25% output, a typical chat or agent mix; input and output are shown separately too. Prices are OpenRouter's, which pass through each provider's list price for most models. Checked twice a day.

| Model | Maker | 1M tokens | Input /1M | Output /1M | Context | 7-day change |
|---|---|---|---|---|---|---|
| Ling 3.0 Tiny | inclusionai | **$0.022** | $0.015 | $0.045 | 262k |  |
| Mistral Nemo | mistralai | **$0.029** | $0.029 | $0.030 | 131k |  |
| Extract | scaledown | **$0.030** | $0.040 | $0.00 | 1,000k |  |
| Summarize | scaledown | **$0.030** | $0.040 | $0.00 | 1,000k |  |
| Ling 3.0 Flash | inclusionai | **$0.032** | $0.021 | $0.063 | 262k |  |
| Llama 3 8B Lunaris | sao10k | **$0.042** | $0.040 | $0.050 | 8k |  |
| Nex-N2.5-Mini | nex-agi | **$0.044** | $0.025 | $0.10 | 262k |  |
| GLM Flash Latest | ~z-ai | **$0.044** | $0.028 | $0.092 | 1,048k |  |
| gpt-oss-120b | openai | **$0.045** | $0.027 | $0.098 | 131k |  |
| gpt-oss-20b | openai | **$0.055** | $0.030 | $0.13 | 131k |  |
| Qwen3.7 Flash | qwen | **$0.055** | $0.030 | $0.13 | 1,000k |  |
| Llama 3.1 8B Instruct | meta-llama | **$0.058** | $0.050 | $0.080 | 131k |  |
| Mistral Small 3 | mistralai | **$0.058** | $0.050 | $0.080 | 32k |  |
| Schematron V2 Turbo | inference-net | **$0.060** | $0.030 | $0.15 | 128k |  |
| Nova Micro 1.0 | amazon | **$0.061** | $0.035 | $0.14 | 128k |  |
| Gemma 3 4B | google | **$0.062** | $0.050 | $0.10 | 131k |  |
| Command R7B (12-2024) | cohere | **$0.066** | $0.037 | $0.15 | 128k |  |
| Mercury 2.5 | inception | **$0.068** | $0.040 | $0.15 | 260k |  |
| Llama 3.2 1B Instruct | meta-llama | **$0.071** | $0.027 | $0.20 | 60k |  |
| Gemma 3 12B | google | **$0.075** | $0.050 | $0.15 | 131k |  |
| Hy-MT2-1.8B | tencent | **$0.077** | $0.044 | $0.18 | 8k |  |
| MythoMax 13B | gryphe | **$0.087** | $0.080 | $0.11 | 8k |  |
| Phi 4 | microsoft | **$0.088** | $0.070 | $0.14 | 16k |  |
| Solar Mini 4 | upstage | **$0.088** | $0.050 | $0.20 | 524k |  |
| Ling 3.0 Flash VL | inclusionai | **$0.090** | $0.060 | $0.18 | 131k |  |
| Ternary Bonsai 2 9B | prism-ml | **$0.092** | $0.080 | $0.13 | 262k |  |
| Schematron V2 Small | inference-net | **$0.095** | $0.050 | $0.23 | 128k |  |
| Ministral 3 3B 2512 | mistralai | **$0.10** | $0.10 | $0.10 | 131k |  |
| Reka Edge | rekaai | **$0.10** | $0.10 | $0.10 | 16k |  |
| Nemotron 3.5 Lightning | nvidia | **$0.10** | $0.070 | $0.20 | 262k |  |
| Nova Lite 1.0 | amazon | **$0.10** | $0.060 | $0.24 | 300k |  |
| Nemotron 3 Nano 30B A3B | nvidia | **$0.10** | $0.060 | $0.24 | 262k |  |
| Mistral Small 3.2 24B | mistralai | **$0.11** | $0.075 | $0.20 | 256k |  |
| Granite 4.2 8B | ibm-granite | **$0.11** | $0.060 | $0.25 | 131k |  |
| Ling 3.1 Flash | inclusionai | **$0.11** | $0.075 | $0.23 | 1,000k |  |
| DeepSeek V4 Flash 0423 | deepseek | **$0.11** | $0.091 | $0.18 | 1,048k |  |
| Qwen3.5-Flash | qwen | **$0.11** | $0.065 | $0.26 | 1,000k |  |
| Nex-N2.5-Pro | nex-agi | **$0.12** | $0.075 | $0.25 | 262k |  |
| DeepSeek V4 Flash 0731 | deepseek | **$0.12** | $0.059 | $0.30 | 1,048k |  |
| DeepSeek V4 Flash Latest | ~deepseek | **$0.12** | $0.059 | $0.30 | 1,048k |  |
| Llama 3.2 3B Instruct | meta-llama | **$0.12** | $0.051 | $0.34 | 131k |  |
| UI-TARS 7B | bytedance | **$0.12** | $0.10 | $0.20 | 128k |  |
| Muse Spark 1.2 Contributor | meta | **$0.12** | $0.10 | $0.20 | 1,048k |  |
| Muse Spark 1.3 Contributor | meta | **$0.12** | $0.10 | $0.20 | 1,048k |  |
| Qwen2.5 7B Instruct | qwen | **$0.12** | $0.10 | $0.20 | 32k |  |
| Reka Flash 3 | rekaai | **$0.12** | $0.10 | $0.20 | 65k |  |
| Hy-MT2-30B-A3B | tencent | **$0.13** | $0.074 | $0.29 | 8k |  |
| Hy-MT2-7B | tencent | **$0.13** | $0.074 | $0.29 | 8k |  |
| DeepSeek Flash Latest | ~deepseek | **$0.13** | $0.006 | $0.50 | 1,048k |  |
| Qwen3 32B | qwen | **$0.13** | $0.080 | $0.28 | 131k |  |
| Seed 1.6 Flash | bytedance-seed | **$0.13** | $0.075 | $0.30 | 262k |  |
| gpt-oss-safeguard-20b | openai | **$0.13** | $0.075 | $0.30 | 131k |  |
| GPT-5 Nano | openai | **$0.14** | $0.050 | $0.40 | 400k |  |
| Gemma 4 26B A4B | google | **$0.14** | $0.090 | $0.30 | 262k |  |
| GLM 4.7 Flash | z-ai | **$0.15** | $0.060 | $0.40 | 200k |  |
| Gemma 3 27B | google | **$0.15** | $0.080 | $0.35 | 131k |  |
| Ministral 3 8B 2512 | mistralai | **$0.15** | $0.15 | $0.15 | 262k |  |
| Qwen3 14B | qwen | **$0.15** | $0.12 | $0.24 | 40k |  |
| Llama 4 Scout | meta-llama | **$0.15** | $0.10 | $0.30 | 1,310k |  |
| Voxtral Small 24B 2507 | mistralai | **$0.15** | $0.10 | $0.30 | 32k |  |
| Qwen3 30B A3B Instruct 2507 | qwen | **$0.15** | $0.10 | $0.30 | 262k |  |
| Step 3.5 Flash | stepfun | **$0.15** | $0.10 | $0.30 | 262k |  |
| Nemotron 3 Super | nvidia | **$0.16** | $0.075 | $0.40 | 1,048k |  |
| Mercury 3 | inception | **$0.16** | $0.10 | $0.35 | 1,000k |  |
| Particle-1 | particle-ai | **$0.16** | $0.050 | $0.50 | 1,048k |  |
| Seed-2.0-Mini | bytedance-seed | **$0.18** | $0.10 | $0.40 | 262k |  |
| Gemini 2.5 Flash Lite | google | **$0.18** | $0.10 | $0.40 | 1,048k |  |
| GPT-4.1 Nano | openai | **$0.18** | $0.10 | $0.40 | 1,047k |  |
| MiMo-V2.6-Flash | xiaomi | **$0.18** | $0.14 | $0.28 | 1,050k |  |
| DeepSeek V4.1 Flash | deepseek | **$0.18** | $0.051 | $0.57 | 1,048k |  |
| Llama Guard 4 12B | meta-llama | **$0.18** | $0.18 | $0.18 | 163k |  |
| Ternary Bonsai 2 27B | prism-ml | **$0.18** | $0.075 | $0.50 | 262k |  |
| Qwen3.5-9B | qwen | **$0.19** | $0.17 | $0.25 | 262k |  |
| Claude Haiku 5.5 | anthropic | **$0.20** | $0.10 | $0.50 | 1,000k |  |
| Ministral 3 14B 2512 | mistralai | **$0.20** | $0.20 | $0.20 | 262k |  |
| Nemotron 3.5 Content Safety | nvidia | **$0.20** | $0.20 | $0.20 | 131k |  |
| GPT-6 Luna | openai | **$0.20** | $0.10 | $0.50 | 1,050k |  |
| GPT-6 Luna Pro | openai | **$0.20** | $0.10 | $0.50 | 1,050k |  |
| Claude Haiku Latest | ~anthropic | **$0.20** | $0.10 | $0.50 | 1,000k |  |
| GPT Luna Latest | ~openai | **$0.20** | $0.10 | $0.50 | 1,050k |  |
| LFM2.5-24B-A2B | liquid | **$0.21** | $0.050 | $0.70 | 128k |  |
| Qwen3 30B A3B | qwen | **$0.21** | $0.12 | $0.50 | 40k |  |
| Qwen3.8 Flash | qwen | **$0.23** | $0.15 | $0.47 | 1,000k |  |
| Qwen3.8 Omni Flash | qwen | **$0.23** | $0.15 | $0.47 | 1,000k |  |
| Hy3 | tencent | **$0.23** | $0.13 | $0.53 | 262k |  |
| Nemotron 3.5 Super VL | nvidia | **$0.24** | $0.15 | $0.50 | 1,048k |  |
| GLM 5.3 Flash | z-ai | **$0.24** | $0.15 | $0.50 | 1,048k |  |
| Hunyuan A13B Instruct | tencent | **$0.25** | $0.14 | $0.57 | 131k |  |
| Gemma 4 31B | google | **$0.25** | $0.20 | $0.40 | 262k |  |
| Command R (08-2024) | cohere | **$0.26** | $0.15 | $0.60 | 128k |  |
| DeepSeek V4.1 Flash 1017 | deepseek | **$0.26** | $0.15 | $0.60 | 1,048k |  |
| MiniMax M3.1 Flash | minimax | **$0.26** | $0.15 | $0.60 | 1,048k |  |
| Mistral Small 4 | mistralai | **$0.26** | $0.15 | $0.60 | 262k |  |
| GPT-4o-mini | openai | **$0.26** | $0.15 | $0.60 | 128k |  |
| GPT-4o-mini (2024-07-18) | openai | **$0.26** | $0.15 | $0.60 | 128k |  |
| Qwen3 Coder 30B A3B Instruct | qwen | **$0.26** | $0.15 | $0.60 | 262k |  |
| Qwen3 VL 30B A3B Instruct | qwen | **$0.26** | $0.15 | $0.60 | 262k |  |
| Solar Pro 4 | upstage | **$0.26** | $0.15 | $0.60 | 524k |  |
| Hy3 preview | tencent | **$0.29** | $0.18 | $0.60 | 262k |  |
| DeepSeek V3.2 | deepseek | **$0.29** | $0.26 | $0.38 | 163k |  |
| Qwen3 Coder Next | qwen | **$0.29** | $0.12 | $0.80 | 262k |  |
| Saba | mistralai | **$0.30** | $0.20 | $0.60 | 32k |  |
| Qwen3.6 35B A3B | qwen | **$0.30** | $0.10 | $0.90 | 262k |  |
| ricursiveAI/Ripple-65B-Preview | ricursiveai | **$0.30** | $0.20 | $0.60 | 1,048k |  |
| DeepSeek V3.2 Exp | deepseek | **$0.30** | $0.27 | $0.41 | 163k |  |
| Qwen3 235B A22B Instruct 2507 | qwen | **$0.31** | $0.14 | $0.80 | 262k |  |
| GLM 4.5 Air | z-ai | **$0.31** | $0.13 | $0.85 | 131k |  |
| Qwen3 Next 80B A3B Instruct | qwen | **$0.34** | $0.090 | $1.10 | 262k |  |
| Cydonia 24B V4.1 | thedrummer | **$0.35** | $0.30 | $0.50 | 131k |  |
| Qwen3.5-35B-A3B | qwen | **$0.35** | $0.14 | $1.00 | 262k |  |
| MiniMax M2.7 | minimax | **$0.37** | $0.21 | $0.84 | 204k |  |
| Qwen2.5 72B Instruct | qwen | **$0.37** | $0.36 | $0.40 | 32k |  |
| Uncensored | cognitivecomputations | **$0.38** | $0.20 | $0.90 | 128k |  |
| Mercury 2 | inception | **$0.38** | $0.25 | $0.75 | 128k |  |
| Qwen3 VL 8B Instruct | qwen | **$0.38** | $0.25 | $0.75 | 262k |  |
| LFM2.5-2.6B | liquid | **$0.39** | $0.35 | $0.50 | 128k |  |
| Qwen-Plus | qwen | **$0.39** | $0.26 | $0.78 | 1,000k |  |
| Qwen3 Coder Flash | qwen | **$0.39** | $0.20 | $0.97 | 1,000k |  |
| UnslopNemo 12B | thedrummer | **$0.40** | $0.40 | $0.40 | 1,024k |  |
| Mistral Small 3.1 24B | mistralai | **$0.40** | $0.35 | $0.56 | 128k |  |
| GLM 5.3 | z-ai | **$0.40** | $0.070 | $1.40 | 1,048k |  |
| GLM Latest | ~z-ai | **$0.40** | $0.070 | $1.40 | 1,048k |  |
| Qwen3.6 Flash | qwen | **$0.42** | $0.19 | $1.12 | 1,000k |  |
| ReMM SLERP 13B | undi95 | **$0.42** | $0.35 | $0.65 | 6k |  |
| DeepSeek V3.1 | deepseek | **$0.42** | $0.25 | $0.95 | 163k |  |
| Step 3.7 Flash | stepfun | **$0.44** | $0.20 | $1.15 | 262k |  |
| MiniMax M2.5 | minimax | **$0.44** | $0.27 | $0.95 | 204k |  |
| Codestral 2508 | mistralai | **$0.45** | $0.30 | $0.90 | 256k |  |
| GLM 4.6V | z-ai | **$0.45** | $0.30 | $0.90 | 131k |  |
| GPT-5.6 Luna | openai | **$0.45** | $0.20 | $1.20 | 1,050k |  |
| GPT-5.6 Luna Pro | openai | **$0.45** | $0.20 | $1.20 | 1,050k |  |
| DeepSeek V3.1 Terminus | deepseek | **$0.45** | $0.27 | $1.00 | 163k |  |
| DeepSeek V3 | deepseek | **$0.46** | $0.32 | $0.89 | 163k |  |
| GPT-5.4 Nano | openai | **$0.46** | $0.20 | $1.25 | 400k |  |
| Qwen3 VL 30B A3B Thinking | qwen | **$0.47** | $0.29 | $1.00 | 262k |  |
| Qwen3 Coder 480B A35B | qwen | **$0.47** | $0.30 | $1.00 | 262k |  |
| Beam 1.0 501B A23B | reflection | **$0.47** | $0.30 | $1.00 | 1,048k |  |
| Perceptron Mk1 | perceptron | **$0.49** | $0.15 | $1.50 | 32k |  |
| Perceptron Mk1.5 | perceptron | **$0.49** | $0.15 | $1.50 | 36k |  |
| Weaver (alpha) | mancer | **$0.49** | $0.40 | $0.75 | 8k |  |
| DeepSeek V3 0324 | deepseek | **$0.50** | $0.29 | $1.14 | 163k |  |
| Llama 4 Maverick | meta-llama | **$0.51** | $0.35 | $1.00 | 524k |  |
| LongCat 2.0 | meituan | **$0.52** | $0.30 | $1.20 | 1,048k |  |
| Muse Glimmer 30B | meta | **$0.52** | $0.30 | $1.20 | 131k |  |
| MiniMax M3 | minimax | **$0.52** | $0.30 | $1.20 | 1,048k |  |
| MiMo-V2.6-Pro | xiaomi | **$0.54** | $0.43 | $0.87 | 1,050k |  |
| Qwen3.7 Plus | qwen | **$0.56** | $0.32 | $1.28 | 1,000k |  |
| Gemini 3.1 Flash Lite | google | **$0.56** | $0.25 | $1.50 | 1,048k |  |
| Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | google | **$0.56** | $0.25 | $1.50 | 65k |  |
| Gemini 3.1 Flash Lite Preview | google | **$0.56** | $0.25 | $1.50 | 1,048k |  |

US dollars as listed on 2026-11-06; the 150 cheapest of 357 paid models shown. Free and variable-priced routes are left out. Some models charge more above a prompt size (often 200k tokens); the base rate is shown. Batch and cached-input discounts are not included.

---
As of 2026-11-06. Canonical page: https://sharpest.ai/cloud/llm-tokens