How accurate is the token count?
It is a close estimate. Each provider uses its own tokenizer, so a single count cannot match all of them exactly. It tracks English prose closely and drifts most on code and non Latin scripts.
Count the tokens. See the real price.
Paste any prompt, document, or transcript. Whizi counts the tokens and prices it across 100 models, side by side.
How the token count works
Every provider ships its own tokenizer, so one number cannot be exact for all of them at once. This counter tracks common English prose closely and drifts furthest on code, long digit runs, and non Latin scripts. Use it to size and compare a prompt, not to reconcile an invoice.
All model prices
100 models, in USD per million tokens, from the OpenRouter catalogue on August 20, 2026. List rates, before any cache discount or volume agreement.
| # | Model | Provider | $ / 1M in | $ / 1M out | Context |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 5 | Anthropic | $2.00 | $10.00 | 1M |
| 2 | Gemini 3.5 Flash | $1.50 | $9.00 | 1M | |
| 3 | GPT-5.5 | OpenAI | $5.00 | $30.00 | 1M |
| 4 | Claude Opus 5 | Anthropic | $5.00 | $25.00 | 1M |
| 5 | Grok 4.5 | xAI | $2.00 | $6.00 | 500K |
| 6 | DeepSeek V3.2 | DeepSeek | $0.27 | $0.40 | 164K |
| 7 | GPT-5.4 Mini | OpenAI | $0.75 | $4.50 | 400K |
| 8 | Gemini 3.1 Pro Preview | $2.00 | $12.00 | 1M | |
| 9 | Qwen3 Coder Next | Qwen | $0.12 | $0.80 | 262K |
| 10 | Kimi K2 Thinking | Moonshot | $0.60 | $2.50 | 262K |
| 11 | GLM 4.6 | Z.ai | $0.50 | $2.00 | 205K |
| 12 | Claude Haiku 4.5 | Anthropic | $1.00 | $5.00 | 200K |
| 13 | GPT-5.4 | OpenAI | $2.50 | $15.00 | 1M |
| 14 | Gemini 2.5 Flash | $0.30 | $2.50 | 1M | |
| 15 | MiniMax M2 | MiniMax | $0.26 | $1.02 | 205K |
| 16 | Claude Sonnet 4.6 | Anthropic | $3.00 | $15.00 | 1M |
| 17 | DeepSeek V3.1 | DeepSeek | $0.25 | $0.95 | 164K |
| 18 | Llama 4 Maverick | Meta | $0.20 | $0.80 | 1M |
| 19 | GPT-5.4 Nano | OpenAI | $0.20 | $1.25 | 400K |
| 20 | Gemini 2.5 Flash Lite | $0.10 | $0.40 | 1M | |
| 21 | Mistral Medium 3.1 | Mistral | $0.40 | $2.00 | 131K |
| 22 | Qwen3.8 Max | Qwen | $2.00 | $6.00 | 1M |
| 23 | GLM 5 | Z.ai | $0.60 | $1.92 | 205K |
| 24 | Claude Sonnet 4.5 | Anthropic | $3.00 | $15.00 | 1M |
| 25 | Kimi K2 0711 | Moonshot | $0.57 | $2.30 | 131K |
| 26 | Grok 4.3 | xAI | $1.25 | $2.50 | 1M |
| 27 | R1 | DeepSeek | $0.70 | $2.50 | 64K |
| 28 | Gemini 3.6 Flash | $0.75 | $3.75 | 1M | |
| 29 | Codestral 2508 | Mistral | $0.30 | $0.90 | 256K |
| 30 | Llama 3.3 70B Instruct | Meta | $0.10 | $0.32 | 131K |
| 31 | GPT Chat Latest | OpenAI | $5.00 | $30.00 | 400K |
| 32 | Mistral Large 3 2512 | Mistral | $0.50 | $1.50 | 262K |
| 33 | MiniMax M3 | MiniMax | $0.30 | $1.20 | 1M |
| 34 | Qwen3.6 Flash | Qwen | $0.19 | $1.13 | 1M |
| 35 | GLM 4.7 | Z.ai | $0.40 | $1.75 | 205K |
| 36 | Kimi K3 | Moonshot | $3.00 | $15.00 | 1M |
| 37 | Claude Opus 4.8 | Anthropic | $5.00 | $25.00 | 1M |
| 38 | Command A | Cohere | $2.50 | $10.00 | 256K |
| 39 | Nova Pro 1.0 | Amazon | $0.80 | $3.20 | 300K |
| 40 | Phi 4 | Microsoft | $0.07 | $0.14 | 16K |
| 41 | Hy-MT2-1.8B | Tencent | $0.04 | $0.18 | 8K |
| 42 | Hy-MT2-30B-A3B | Tencent | $0.07 | $0.29 | 8K |
| 43 | GLM 5.3 | Z.ai | $1.40 | $4.40 | 1M |
| 44 | Qwen3.8 27B | Qwen | $0.45 | $3.20 | 1M |
| 45 | Gemini 3.7 Flash | $0.38 | $1.88 | 1M | |
| 46 | Seed 2.1 Turbo | Bytedance Seed | $0.50 | $2.50 | 262K |
| 47 | Qwen3.8 2.4T A95B | Qwen | $2.00 | $6.00 | 1M |
| 48 | Seed-2.0-Code | Bytedance Seed | $0.50 | $3.00 | 262K |
| 49 | DeepSeek V4 Pro 0813 | DeepSeek | $1.19 | $3.56 | 1M |
| 50 | Grok 4.6 | xAI | $2.00 | $6.00 | 500K |
| 51 | Nemotron 3.5 Lightning | NVIDIA | $0.08 | $0.20 | 262K |
| 52 | Sakana Namazu | Sakana | $0.95 | $4.00 | 262K |
| 53 | Solar Pro 4 | Upstage | $0.03 | $0.12 | 524K |
| 54 | Muse Glimmer 30B | Meta | $0.35 | $1.50 | 131K |
| 55 | Muse Spark 1.2 | Meta | $1.25 | $4.25 | 1M |
| 56 | DeepSeek V4 Flash 0731 | DeepSeek | $0.14 | $0.28 | 1M |
| 57 | Inkling Small | Thinkingmachines | $0.45 | $1.20 | 524K |
| 58 | Qwen3.7 Flash | Qwen | $0.03 | $0.13 | 1M |
| 59 | Claude Opus 5 (Fast) | Anthropic | $10.00 | $50.00 | 1M |
| 60 | Ling-3.0-flash | inclusionAI | $0.02 | $0.06 | 262K |
| 61 | Laguna S 2.1 | Poolside | $0.09 | $0.18 | 1M |
| 62 | Gemini 3.5 Flash Lite | $0.30 | $2.50 | 1M | |
| 63 | LongCat 2.0 | Meituan | $0.30 | $1.20 | 1M |
| 64 | Inkling | Thinkingmachines | $0.95 | $4.05 | 1M |
| 65 | Muse Spark 1.1 | Meta | $1.25 | $4.25 | 1M |
| 66 | KAT-Coder-Air V2.5 | Kwaipilot | $0.15 | $0.60 | 256K |
| 67 | KAT-Coder-Pro V2.5 | Kwaipilot | $0.74 | $2.96 | 256K |
| 68 | GPT-5.6 Luna Pro | OpenAI | $0.20 | $1.20 | 1M |
| 69 | GPT-5.6 Luna | OpenAI | $0.20 | $1.20 | 1M |
| 70 | GPT-5.6 Terra Pro | OpenAI | $2.00 | $12.00 | 1M |
| 71 | GPT-5.6 Terra | OpenAI | $2.00 | $12.00 | 1M |
| 72 | GPT-5.6 Sol Pro | OpenAI | $2.50 | $15.00 | 1M |
| 73 | GPT-5.6 Sol | OpenAI | $2.50 | $15.00 | 1M |
| 74 | Aion-3.0-Mini | Aion Labs | $0.70 | $1.40 | 131K |
| 75 | Aion-3.0 | Aion Labs | $3.00 | $6.00 | 131K |
| 76 | Hy3 | Tencent | $0.13 | $0.53 | 262K |
| 77 | Laguna XS 2.1 | Poolside | $0.06 | $0.12 | 262K |
| 78 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | $0.25 | $1.50 | 66K | |
| 79 | Nex-N2-Mini | Nex Agi | $0.03 | $0.10 | 262K |
| 80 | Fugu Ultra | Sakana | $5.00 | $30.00 | 1M |
| 81 | Nano Banana 2 (Gemini 3.1 Flash Image) | $0.50 | $3.00 | 131K | |
| 82 | Nano Banana Pro (Gemini 3 Pro Image) | $2.00 | $12.00 | 131K | |
| 83 | GLM 5.2 | Z.ai | $0.97 | $3.04 | 1M |
| 84 | Kimi K2.7 Code | Moonshot | $0.71 | $3.50 | 262K |
| 85 | Claude Fable 5 | Anthropic | $10.00 | $50.00 | 1M |
| 86 | Nex-N2-Pro | Nex Agi | $0.25 | $1.00 | 262K |
| 87 | Nemotron 3 Ultra | NVIDIA | $0.60 | $3.60 | 512K |
| 88 | Qwen3.7 Plus | Qwen | $0.32 | $1.28 | 1M |
| 89 | Step 3.7 Flash | Stepfun | $0.20 | $1.15 | 262K |
| 90 | Claude Opus 4.8 (Fast) | Anthropic | $10.00 | $50.00 | 1M |
| 91 | Qwen3.7 Max | Qwen | $1.48 | $4.42 | 1M |
| 92 | Grok Build 0.1 | xAI | $1.00 | $2.00 | 256K |
| 93 | Claude Opus 4.7 (Fast) | Anthropic | $30.00 | $150.00 | 1M |
| 94 | Perceptron Mk1 | Perceptron | $0.15 | $1.50 | 33K |
| 95 | Ring-2.6-1T | inclusionAI | $0.07 | $0.63 | 262K |
| 96 | Gemini 3.1 Flash Lite | $0.25 | $1.50 | 1M | |
| 97 | Granite 4.1 8B | Ibm Granite | $0.05 | $0.10 | 131K |
| 98 | Mistral Medium 3.5 | Mistral | $1.50 | $7.50 | 262K |
| 99 | Qwen3.5 Plus 2026-04-20 | Qwen | $0.30 | $1.80 | 1M |
| 100 | Qwen3.6 35B A3B | Qwen | $0.14 | $1.00 | 262K |
Common questions
Where do the prices come from?
From the public OpenRouter model catalogue, in USD per million tokens. They are list rates and do not include prompt caching discounts, batch pricing, or negotiated volume rates.
Why does output cost more than input?
Generating tokens takes far more compute than reading them, so most providers charge two to five times more for output. A short prompt with a long reply can cost more than a long prompt with a short reply.
Next step
Compare plans on Whizi pricing or estimate your current stack with the AI subscription savings calculator.