rates verified 2026-09-02
API token pricing
All 192 models across 22 providers, priced per 1M tokens. Tap a column to sort; cell color runs from cheap (green) to expensive (orange) on a log scale. Prices come from each provider's official price list.
| Tier | |||||
|---|---|---|---|---|---|
GPT-5.6 Sol | Frontier | $4 | $0.40 | $20 | 1.1M |
| Flat plan | $2.80 | $0.28 | $14 | Try → | |
| Host | $5 | $0.50 | $30 | 1.1M | |
GPT-5.6 Terra | Mid | $2 | $0.20 | $12 | 1.1M |
| Flat plan | $1.40 | $0.14 | $8.40 | Try → | |
| Host | $2 | $0.20 | $12 | 1.1M | |
GPT-5.6 Luna | Budget | $0.20 | $0.02 | $1.20 | 1.1M |
| Flat plan | $0.14 | $0.01 | $0.84 | Try → | |
| Host | $0.20 | $0.02 | $1.20 | 1.1M | |
GPT-5.5 | Frontier | $5 | $0.50 | $30 | — |
| Host | $5 | $0.50 | $30 | 1.1M | |
GPT-5.5 Pro | Frontier | $30 | — | $180 | — |
Claude Fable 5 | Frontier | $10 | $1 | $50 | 1M |
Claude Opus 5 | Frontier | $5 | $0.50 | $25 | 1M |
Claude Sonnet 5 | Mid | $2 | $0.20 | $10 | 1M |
Claude Haiku 4.5 | Budget | $1 | $0.10 | $5 | 200K |
Gemini 3.7 Flash | Mid | $0.75 | $0.07 | $3.75 | — |
Gemini 3.6 Flash | Mid | $0.75 | $0.07 | $3.75 | — |
Gemini 3.5 Flash | Mid | $1.50 | $0.15 | $9 | — |
Gemini 3.5 Flash-Lite | Budget | $0.30 | $0.03 | $2.50 | — |
Gemini 3.1 Flash-Lite | Budget | $0.25 | $0.03 | $1.50 | — |
Gemini 3.1 Pro Previewpreview | Frontier | $2 | $0.20 | $12 | — |
DeepSeek V4 Flash | Budget | $0.44 | $0.01 | $1.32 | 1M |
viaDeepInfra | Host | $0.08 | $0.02 | $0.18 | 1.0M |
viaNebius | Host | $0.14 | — | $0.28 | 1.0M |
viaNovita AI | Host | $0.14 | — | $0.28 | 1.0M |
viaTogether AI | Host | $0.14 | — | $0.28 | 1M |
viaFireworks AI | Host | $0.22 | $0.0070 | $0.66 | — |
DeepSeek V4 Flash Vision (Experimental)preview | Budget | $0.44 | $0.01 | $1.32 | 1M |
DeepSeek V4 Pro | Frontier | $1.32 | $0.04 | $3.96 | 1M |
viaDeepInfra | Host | $1.30 | $0.10 | $2.60 | 1.0M |
viaFireworks AI | Host | $1.32 | $0.04 | $3.96 | — |
viaNovita AI | Host | $1.32 | — | $3.96 | 1.0M |
viaTogether AI | Host | $1.32 | — | $3.96 | 1.0M |
viaNebius | Host | $1.75 | — | $3.50 | — |
Grok 4.6 | Frontier | $2 | $0.50 | $6 | 500K |
Grok 4.5 | Frontier | $2 | $0.30 | $6 | 500K |
Grok 4.3 | Mid | $1.25 | $0.20 | $2.50 | 1M |
Grok 4.20 Reasoning (0309) | Mid | $1.25 | $0.20 | $2.50 | 1M |
Grok 4.20 Non-Reasoning (0309) | Mid | $1.25 | $0.20 | $2.50 | 1M |
Grok 4.20 Multi-Agent (0309) | Mid | $1.25 | $0.20 | $2.50 | 1M |
Grok Build 0.1 | Budget | $1 | $0.20 | $2 | 256K |
Kimi K3 | Frontier | $3 | $0.30 | $15 | 1.0M |
viaDeepInfra | Host | $2.85 | $0.28 | $14.25 | 1.0M |
viaFireworks AI | Host | $3 | $0.30 | $15 | — |
viaNebius | Host | $3 | — | $15 | 1.0M |
viaNovita AI | Host | $3 | — | $15 | 1.0M |
viaTogether AI | Host | $3 | — | $15 | 1.0M |
Kimi K2.7 Code High-Speed | Mid | $1.90 | $0.38 | $8 | 262K |
Kimi K2.7 Code | Mid | $0.95 | $0.19 | $4 | 262K |
viaDeepInfra | Host | $0.68 | $0.14 | $3.40 | 262K |
viaFireworks AI | Host | $0.95 | $0.19 | $4 | — |
viaNebius | Host | $0.95 | — | $4 | — |
Kimi K2.6 | Mid | $0.95 | $0.16 | $4 | 262K |
viaDeepInfra | Host | $0.75 | $0.15 | $3.50 | 262K |
viaFireworks AI | Host | $0.95 | $0.16 | $4 | — |
viaNebius | Host | $0.95 | — | $4 | — |
Qwen3.8 Max | Frontier | $2 | — | $6 | 1M |
viaFireworks AI | Host | $2 | $0.25 | $6 | — |
Qwen3.7 Max | Frontier | $2.50 | — | $7.50 | 1M |
viaTogether AI | Host | $1.25 | — | $3.75 | — |
Qwen3 Coder Plus | Mid | $1 | — | $5 | 1M |
Qwen3 Coder Flash | Budget | $0.30 | — | $1.50 | 1M |
Qwen Plus | Mid | $0.40 | — | $1.20 | 1M |
Qwen Flash | Budget | $0.05 | — | $0.40 | 1M |
Qwen Turbo | Budget | $0.05 | — | $0.20 | — |
Mistral Small 4 | Budget | $0.15 | — | $0.60 | — |
Mistral Medium 3.5 | Frontier | $1.50 | — | $7.50 | — |
Mistral Large 3 | Frontier | $0.50 | — | $1.50 | — |
Ministral 3 14B | Budget | $0.20 | — | $0.20 | — |
Ministral 3 8B | Budget | $0.15 | — | $0.15 | — |
Ministral 3 3B | Budget | $0.10 | — | $0.10 | — |
Codestral | Mid | $0.30 | — | $0.90 | — |
GLM-5.3 | Frontier | $1.40 | $0.26 | $4.40 | 1M |
viaFireworks AI | Host | $1.40 | $0.26 | $4.40 | — |
viaTogether AI | Host | $1.40 | — | $4.40 | 1M |
GLM-5.3-Flash | Budget | $0.07 | $0.01 | $0.25 | — |
viaNovita AI | Host | $0.07 | — | $0.25 | 1.0M |
viaFireworks AI | Host | $0.15 | $0.03 | $0.50 | — |
viaNebius | Host | $0.15 | — | $0.50 | — |
viaTogether AI | Host | $0.15 | — | $0.50 | 1M |
GLM-5.2 | Frontier | $1.40 | $0.26 | $4.40 | 1M |
viaFireworks AI | Host | $1.40 | $0.14 | $4.40 | — |
viaNebius | Host | $1.40 | — | $4.40 | — |
GLM-5.1 | Frontier | $1.40 | $0.26 | $4.40 | — |
GLM-5 | Mid | $1 | $0.20 | $3.20 | — |
GLM-4.7 | Mid | $0.60 | $0.11 | $2.20 | 200K |
GLM-4.7-FlashX | Budget | $0.07 | $0.01 | $0.40 | 200K |
Sonar Deep Research | Frontier | $2 | — | $8 | 128K |
Sonar | Budget | $1 | — | $1 | 128K |
Sonar Pro | Frontier | $3 | — | $15 | 200K |
Sonar Reasoning Pro | Mid | $2 | — | $8 | 128K |
Command R7B | Budget | $0.04 | — | $0.15 | 128K |
Command R | Mid | $0.15 | — | $0.60 | 128K |
MiniMax M3 | Frontier | $0.30 | $0.06 | $1.20 | 1M |
viaFireworks AI | Host | $0.30 | $0.06 | $1.20 | — |
viaNebius | Host | $0.30 | — | $1.20 | 1.0M |
viaTogether AI | Host | $0.30 | — | $1.20 | 524K |
MiniMax M2.7 High-Speed | Mid | $0.60 | $0.06 | $2.40 | — |
MiniMax M2.7 | Mid | $0.30 | $0.06 | $1.20 | — |
viaNovita AI | Host | $0.30 | — | $1.20 | 205K |
Nova 2 Lite | Mid | $0.30 | $0.07 | $2.50 | 1M |
Nova 2 Propreview | Frontier | $1.25 | — | $10 | 1M |
Nova Premier | Frontier | $2.50 | — | $12.50 | 1M |
Nova Lite | Budget | $0.06 | $0.01 | $0.24 | 300K |
Nova Micro | Budget | $0.04 | $0.0088 | $0.14 | 128K |
Nova Pro | Mid | $0.80 | $0.20 | $3.20 | 300K |
Jamba Large | Mid | $2 | — | $8 | 256K |
Jamba Mini | Budget | $0.20 | — | $0.40 | 256K |
GPT OSS 120B | Mid | $0.15 | — | $0.60 | 131K |
viaNebius | Host | $0.15 | — | $0.60 | — |
viaTogether AI | Host | $0.15 | — | $0.60 | 128K |
viaCerebras | Host | $0.35 | — | $0.75 | 131K |
Safety GPT OSS 20Bpreview | Budget | $0.07 | — | $0.30 | 131K |
Qwen 3.6-27Bpreview | Mid | $0.60 | — | $3 | 131K |
Llama Prompt Guard 2 86Mpreview | Budget | $0.04 | — | $0.04 | 1K |
Llama Prompt Guard 2 22Mpreview | Budget | $0.03 | — | $0.03 | 1K |
GPT-OSS 20B | Budget | $0.05 | — | $0.20 | 128K |
viaGroq | Host | $0.07 | — | $0.30 | 131K |
Gemma 4 31B Instruct | Mid | $0.39 | — | $0.97 | 262K |
Qwen 3.8-2.4T-A95B | Frontier | $2 | — | $6 | — |
Qwen 3.8 Flash | Budget | $0.15 | — | $0.47 | 1M |
Qwen 3.7 Plus | Mid | $0.32 | — | $1.28 | 1M |
viaFireworks AI | Host | $0.40 | $0.08 | $1.60 | — |
Qwen 3.5 9B | Budget | $0.17 | — | $0.25 | 262K |
Inkling | Frontier | $1 | — | $4.05 | 524K |
NVIDIA Nemotron 3.5 Lightning 30B | Budget | $0.05 | $0.01 | $0.20 | — |
Llama-4-Maverick-17B-128E | Mid | $0.20 | — | $0.80 | 1.0M |
DeepSeek-V4-Flash | Mid | $0.09 | $0.02 | $0.18 | 1.0M |
Qwen3.5-397B-A17B | Frontier | $0.45 | $0.22 | $3 | 262K |
viaNebius | Host | $0.60 | — | $3.60 | — |
Llama-3.3-70B-Instruct-Turbo | Budget | $0.10 | — | $0.32 | 131K |
viaTogether AI | Host | $1.04 | — | $1.04 | 131K |
DeepSeek-V3.2 | Budget | $0.26 | $0.13 | $0.38 | 164K |
viaNovita AI | Host | $0.27 | — | $0.40 | 164K |
Qwen3-Coder-480B-A35B-Instruct-Turbo | Frontier | $0.30 | $0.10 | $1 | 262K |
Qwen3-235B-A22B-Instruct-2507 | Mid | $0.09 | — | $0.55 | 262K |
viaNebius | Host | $0.20 | — | $0.60 | — |
Gemma 4 31B | Mid | $0.99 | — | $1.49 | 131K |
DeepSeek-V4-Flash-0731 | Mid | $0.14 | — | $0.28 | 1.0M |
OpenAI GPT OSS 120B | Budget | $0.05 | — | $0.25 | 131K |
viaFireworks AI | Host | $0.15 | $0.01 | $0.60 | — |
DeepSeek V4 Flash 0731 | Mid | $0.44 | — | $1.32 | 1.0M |
Llama 4 Maverick Instruct | Mid | $0.27 | — | $0.85 | 1.0M |
DeepSeek V4 Pro | Frontier | $1.60 | — | $3.20 | 1.0M |
Qwen3.8 27B | Mid | $0.42 | — | $3 | — |
| Host | $0.80 | — | $4 | 131K |
standardcompute figures are effective flat-plan rates at sustained volume rather than per-token list prices — the methodology page explains how they're derived. All other rows are unmodified official list prices.
Cost of 1M input + 1M output tokens. Bar length is square-root scaled so cheap models stay visible next to premium ones.
Per-token rates only matter multiplied by your traffic. The calculators turn these rates into monthly bills — including cached-input discounts and subscription break-evens.
questions
Frequently asked
›Where do these prices come from?
Every rate is read from the provider's own published price list (the pages linked on each provider profile). We record standard pay-as-you-go rates — not batch, off-peak, or enterprise discounts.
›What does the cached-input column mean?
Most providers charge a reduced rate when a prompt prefix repeats and hits their cache. If your app reuses a long system prompt, the cached rate can matter more than the headline input rate.
›How current is the data?
Rates on this page were last verified on 2026-09-02. Providers change prices without notice, so confirm on the provider's page before committing a budget.
›Can I get this data as JSON?
Yes — the full dataset is available at /api/pricing.json, free to use with attribution. See the API docs page for the schema.
›Why are some models marked legacy?
Legacy models are still purchasable but superseded by a newer generation from the same provider. They're hidden by default; tick the checkbox above the table to include them.