LLM Pricing Calculator
Prices verified on November 11, 2026.
Free LLM pricing calculator. Compare API costs for GPT, Claude, Gemini, DeepSeek & more. Find the cheapest AI model for your workload in seconds.
How LLM pricing works
LLM APIs charge per million tokens, with one price for the input you send and a higher price for the output the model writes.
Pick a model, enter your tokens and press Calculate pricing to see the cost in dollars.
Token cost calculator
- Batch processing halves the price at OpenAI, Anthropic, and Google.
LLM pricing table
Showing 196 text models, newest first. Click a column to sort.
 Gemini 4 Carbon Google | $2.00 | $10.00 | $12.00 |
 Muse Ember Meta | $0.10 | $0.50 | $0.60 |
 Claude Fable 5.2 Anthropic | $8.00 | $40.00 | $48.00 |
 GPT-6.1 Terra OpenAI | $1.00 | $5.00 | $6.00 |
 Kimi K3.1 Pro Moonshot AI (Kimi) | $2.00 | $10.00 | $12.00 |
| Kimi K3.1 Pro Highspeed Moonshot AI (Kimi) | $4.00 | $20.00 | $24.00 |
 DeepSeek V4.1 Flash 1017 DeepSeek Off-peak price. Peak hours cost double | $0.15 | $0.60 | $0.75 |
| Muse Spark Open Meta Price via Ollama Cloud | $1.25 | $4.25 | $5.50 |
| Gemini 4 Argon Google | $2.00 | $10.00 | $12.00 |
| Muse Wonder Meta | $4.00 | $20.00 | $24.00 |
| Claude Haiku 5.5 Anthropic | $0.10 | $0.50 | $0.60 |
 Mistral Large 4 Mistral | $0.70▲ 2.9% | $2.10▲ 0.5% | $2.80 |
| Mistral Large 4.0.8 Mistral | $0.70 | $2.10 | $2.80 |
| GPT-6.1 Sol OpenAI | $2.00 | $10.00 | $12.00 |
| Claude Sonnet 5.5 Anthropic | $2.00 | $10.00 | $12.00 |
 Qwen 3.8 Max Prime Alibaba (Qwen) Price via Vercel AI Gateway | $4.00 | $12.00 | $16.00 |
| GPT-6 Luna OpenAI | $0.10 | $0.50 | $0.60 |
| GPT-6 Sol OpenAI | $2.00 | $10.00 | $12.00 |
| Claude Opus 5.5 Anthropic | $4.00 | $20.00 | $24.00 |
 Grok 4.7 xAI Double price at 200K+ token prompts | $2.00 | $6.00 | $8.00 |
 GLM-5.3-FlashX Z.ai (GLM) | $0.37 | $1.25 | $1.62 |
| Qwen3.8 Omni Flash Alibaba (Qwen) | $0.15 | $0.47 | $0.62 |
| Kimi K2.8 Preview Moonshot AI (Kimi) Price via DevPass (LLM Gateway) | $1.00 | $4.00 | $5.00 |
| DeepSeek V4.1 Flash DeepSeek Price via engy | $0.03▼ 77% | $0.12▼ 77% | $0.15 |
| GPT-6 Astra OpenAI | $10.00 | $50.00 | $60.00 |
| GPT-6 Astra (Fast) OpenAI Price via Vercel AI Gateway | $20.00 | $100.00 | $120.00 |
| Muse Spark 1.3 Contributor Meta | $0.10 | $0.20 | $0.30 |
| Gemini 3.8 Flash Google Intro price until Dec 31, 2026, then doubles | $0.75 | $3.75 | $4.50 |
| Muse Spark 1.3 Meta | $1.25 | $4.25 | $5.50 |
| Qwen3.8 Max 0902 Alibaba (Qwen) Price via Vercel AI Gateway | $2.00 | $6.00 | $8.00 |
| Claude Fable 5.1 Anthropic | $10.00 | $50.00 | $60.00 |
| Qwen3.8 Flash Alibaba (Qwen) | $0.15 | $0.47 | $0.62 |
| GLM-5.3-Flash Z.ai (GLM) | $0.15 | $0.50 | $0.65 |
| Qwen3.8 27B Alibaba (Qwen) Price via engy | $0.04▼ 55% | $0.32▼ 20% | $0.36 |
| GLM-5.3 Z.ai (GLM) | $1.40 | $4.40 | $5.80 |
| DeepSeek V4 Pro DeepSeek Off-peak price. Peak hours cost double | $0.66 | $1.98 | $2.64 |
| DeepSeek V4 Pro 0813 DeepSeek Price via Vercel AI Gateway | $0.66▲ 0.2% | $1.98▲ 0.2% | $2.64 |
| Gemini 3.7 Flash Google | $0.75 | $3.75 | $4.50 |
| Grok 4.6 xAI | $2.00 | $6.00 | $8.00 |
| Qwen3.8 2.4T A95B Alibaba (Qwen) Price via DevPass (LLM Gateway) | $2.00 | $6.00 | $8.00 |
| Muse Glimmer 30B Meta Price via EmpirioLabs AI | $0.20 | $0.80 | $1.00 |
| Muse Spark 1.2 Contributor Meta | $0.10 | $0.20 | $0.30 |
| Muse Spark 1.2 Meta | $1.25 | $4.25 | $5.50 |
| Qwen3.8 Max Alibaba (Qwen) | $2.00 | $6.00 | $8.00 |
| DeepSeek V4 Flash 0731 DeepSeek Price via engy | $0.04▲ 29% | $0.09▲ 29% | $0.14 |
| Claude Opus 5 Anthropic | $5.00 | $25.00 | $30.00 |
| Gemini 3.5 Flash Lite Google | $0.30 | $2.50 | $2.80 |
| Gemini 3.6 Flash Google | $0.75 | $3.75 | $4.50 |
| Kimi K3 Moonshot AI (Kimi) | $3.00 | $15.00 | $18.00 |
| Qwen3.7 Flash Alibaba (Qwen) | $0.03 | $0.13 | $0.16 |
| GPT-5.6 Luna OpenAI | $0.20 | $1.20 | $1.40 |
| GPT-5.6 Terra OpenAI | $2.00 | $12.00 | $14.00 |
| GPT-5.6 OpenAI | $4.00 | $20.00 | $24.00 |
| GPT-5.6 Sol OpenAI Promo price, until at least Nov 21, 2026 | $4.00 | $20.00 | $24.00 |
| Grok 4.5 xAI | $2.00 | $6.00 | $8.00 |
| Claude Sonnet 5 Anthropic | $2.00 | $10.00 | $12.00 |
| GLM-5.2 Z.ai (GLM) | $1.40 | $4.40 | $5.80 |
| Kimi K2.7 Code Moonshot AI (Kimi) | $0.95 | $4.00 | $4.95 |
| Kimi K2.7 Code HighSpeed Moonshot AI (Kimi) | $1.90 | $8.00 | $9.90 |
| Gemma 4 12B IT Google Price via NanoGPT | $0.05 | $0.25 | $0.30 |
| Claude Fable 5 Anthropic | $10.00 | $50.00 | $60.00 |
| Qwen3.7 Plus Alibaba (Qwen) | $0.40 | $1.60 | $2.00 |
| Claude Opus 4.8 Anthropic | $5.00 | $25.00 | $30.00 |
| Qwen3.7 Max Alibaba (Qwen) | $2.50 | $7.50 | $10.00 |
| Gemini 3.5 Flash Google | $1.50 | $9.00 | $10.50 |
| Gemini 3.1 Flash Lite Google | $0.25 | $1.50 | $1.75 |
| Mistral Medium 3.5 Mistral | $1.50 | $7.50 | $9.00 |
| Qwen3.6 Flash Alibaba (Qwen) | $0.25▲ 33% | $1.50▲ 33% | $1.75 |
| DeepSeek V4 Flash 0423 DeepSeek Price via Merge Gateway | $0.14▲ 99% | $0.28▲ 11% | $0.42 |
| DeepSeek V4 Pro 0423 DeepSeek Price via Merge Gateway | $1.65 | $3.30 | $4.95 |
| GPT-5.5 OpenAI | $5.00 | $30.00 | $35.00 |
| GPT-5.5 Pro OpenAI | $30.00 | $180.00 | $210.00 |
| Qwen3.6 27B Alibaba (Qwen) | $0.60 | $3.60 | $4.20 |
| Kimi K2.6 Moonshot AI (Kimi) | $0.95 | $4.00 | $4.95 |
| Qwen3.6 35B-A3B Alibaba (Qwen) | $0.38▲ 51% | $2.25▲ 52% | $2.63 |
| Grok 4.3 xAI | $1.25 | $2.50 | $3.75 |
| Grok Build 0.1 xAI | $1.00 | $2.00 | $3.00 |
| Claude Opus 4.7 Anthropic | $5.00 | $25.00 | $30.00 |
| Muse Spark 1.1 Meta | $1.25 | $4.25 | $5.50 |
| GLM-5.1 Z.ai (GLM) | $1.40 | $4.40 | $5.80 |
| Gemma 4 26B A4B IT Google Price via Kilo Gateway | $0.01▲ 10% | $0.03▲ 13% | $0.04 |
| Gemma 4 31B IT Google Price via DevPass (LLM Gateway) | $0.10▲ 33% | $0.25▲ 98% | $0.35 |
| Qwen3.6 Plus Alibaba (Qwen) | $0.50 | $3.00 | $3.50 |
| GLM-5V-Turbo Z.ai (GLM) | $1.20 | $4.00 | $5.20 |
| GPT-5.4 nano OpenAI | $0.20 | $1.25 | $1.45 |
| GPT-5.4 mini OpenAI | $0.75 | $4.50 | $5.25 |
| Mistral Small 4 Mistral | $0.15 | $0.60 | $0.75 |
| GLM-5-Turbo Z.ai (GLM) | $1.20 | $4.00 | $5.20 |
| Grok 4.20 (Reasoning) xAI | $1.25 | $2.50 | $3.75 |
| Grok 4.20 (Non-Reasoning) xAI | $1.25 | $2.50 | $3.75 |
| Grok 4.20 Multi-Agent xAI | $1.25 | $2.50 | $3.75 |
| GPT-5.4 OpenAI | $2.50 | $15.00 | $17.50 |
| GPT-5.4 Pro OpenAI | $30.00 | $180.00 | $210.00 |
| GPT-5.3 Chat (latest) OpenAI Price via Merge Gateway | $1.75 | $14.00 | $15.75 |
| Qwen3.5 2B Alibaba (Qwen) Price via Requesty | $0.02 | $0.10 | $0.12 |
| Qwen3.5 9B Alibaba (Qwen) Price via DevPass (LLM Gateway) | $0.10▲ 150% | $0.15 | $0.25 |
| Qwen3.5 Flash Alibaba (Qwen) | $0.10 | $0.40 | $0.50 |
| Qwen3.5 35B-A3B Alibaba (Qwen) | $0.25 | $2.00 | $2.25 |
| Qwen3.5 27B Alibaba (Qwen) | $0.30 | $2.40 | $2.70 |
| Qwen3.5 122B-A10B Alibaba (Qwen) | $0.40 | $3.20 | $3.60 |
| Gemini 3.1 Pro Preview Google Higher price above 200K-token prompts | $2.00 | $12.00 | $14.00 |
| Claude Sonnet 4.6 Anthropic | $3.00 | $15.00 | $18.00 |
| Qwen3.5 Plus Alibaba (Qwen) | $0.40 | $2.40 | $2.80 |
| Qwen3.5 397B-A17B Alibaba (Qwen) | $0.60 | $3.60 | $4.20 |
| GLM-5 Z.ai (GLM) | $1.00 | $3.20 | $4.20 |
| GPT-5.3 Codex OpenAI | $1.75 | $14.00 | $15.75 |
| GPT-5.3 Codex Spark OpenAI | $1.75 | $14.00 | $15.75 |
| Claude Opus 4.6 Anthropic | $5.00 | $25.00 | $30.00 |
| Qwen3 Coder Next Alibaba (Qwen) Price via Cortecs | $0.17▲ 55% | $0.89▲ 32% | $1.06 |
| DeepSeek OCR 2 DeepSeek Price via Novita AI | $0.03 | $0.03 | $0.06 |
| Kimi K2.5 Moonshot AI (Kimi) Price via NanoGPT | $0.38 | $2.00 | $2.38 |
| GLM-4.7-FlashX Z.ai (GLM) | $0.07 | $0.40 | $0.47 |
| GLM-4.7-Flash Z.ai (GLM) Price via Cortecs | $0.08▲ 33% | $0.48▲ 20% | $0.56 |
| GLM-4.7 Z.ai (GLM) | $0.60 | $2.20 | $2.80 |
| X-Ai/Grok-4-Fast-Reasoning xAI Price via Helicone | $0.20▲ 11% | $0.50▲ 11% | $0.70 |
| X-Ai/Grok-4-Fast-Non-Reasoning xAI Price via Helicone | $0.20▲ 11% | $0.50▲ 11% | $0.70 |
| Gemini 3 Flash Preview Google | $0.50 | $3.00 | $3.50 |
| GPT-5.2 OpenAI | $1.75 | $14.00 | $15.75 |
| GPT-5.2 Codex OpenAI Price via Azure | $1.75▲ 1,150% | $14.00▲ 1,128% | $15.75 |
| GPT-5.2 Pro OpenAI | $21.00 | $168.00 | $189.00 |
| GLM-4.6V Z.ai (GLM) | $0.30 | $0.90 | $1.20 |
| Ministral 3 3B Mistral | $0.10 | $0.10 | $0.20 |
| Ministral 3 8B Mistral | $0.15 | $0.15 | $0.30 |
| Ministral 14B Mistral Price via Vercel AI Gateway | $0.20 | $0.20 | $0.40 |
| Ministral 3 14B Mistral | $0.20 | $0.20 | $0.40 |
| DeepSeek V3.2 DeepSeek Price via NanoGPT | $0.21▲ 16% | $0.31▼ 12% | $0.52 |
| DeepSeek Chat DeepSeek Price via NanoGPT | $0.10▼ 32% | $0.42▲ 44% | $0.53 |
| DeepSeek Reasoner DeepSeek Price via OrcaRouter | $0.24▲ 65% | $0.73▲ 146% | $0.97 |
| Claude Opus 4.5 Anthropic | $5.00 | $25.00 | $30.00 |
| Gemini 3 Pro Preview Google Price via QiHang | $0.57 | $3.43 | $4.00 |
| GPT-5.1 Codex mini OpenAI Price via Jiekou.AI | $0.24▲ 8.0% | $1.90▲ 5.6% | $2.14 |
| GPT-5.1 Codex OpenAI Price via OpenCode | $1.07 | $8.50 | $9.57 |
| GPT-5.1 Codex Max OpenAI Price via Jiekou.AI | $1.19▲ 8.0% | $9.50▲ 5.6% | $10.69 |
| GPT-5.1 OpenAI | $1.25 | $10.00 | $11.25 |
| Kimi K2 Thinking Moonshot AI (Kimi) Price via Helicone | $0.48 | $2.00 | $2.48 |
| Claude Haiku 4.5 Anthropic | $1.00 | $5.00 | $6.00 |
| GPT-5 Pro OpenAI | $15.00 | $120.00 | $135.00 |
| Qwen3 VL 30B A3B Instruct Alibaba (Qwen) Price via DevPass (LLM Gateway) | $0.15 | $0.60 | $0.75 |
| GLM-4.6 Z.ai (GLM) | $0.60 | $2.20 | $2.80 |
| Gemini 2.5 Flash-Lite Preview 09-2025 Google Price via 302.AI | $0.10 | $0.40 | $0.50 |
| Qwen3 VL 235B A22B Instruct Alibaba (Qwen) Price via DevPass (LLM Gateway) | $0.20 | $0.88 | $1.08 |
| Qwen3-VL Plus Alibaba (Qwen) | $0.20 | $1.60 | $1.80 |
| GPT-5-Codex OpenAI Price via OpenCode | $1.07 | $8.50 | $9.57 |
| Kimi K2 Instruct 0905 Moonshot AI (Kimi) Price via Helicone | $0.50 | $2.00 | $2.50 |
| x-AI/Grok-Code-Fast 1 xAI Price via Jiekou.AI | $0.19▲ 5.6% | $1.43▲ 5.6% | $1.61 |
| DeepSeek-V3.1 DeepSeek Price via TokenGo | $0.19 | $0.71 | $0.90 |
| DeepSeek V3.1 DeepSeek Price via OpenRouter | $0.25 | $0.95 | $1.20 |
| GLM-4.5V Z.ai (GLM) | $0.60 | $1.80 | $2.40 |
| GPT-5 Nano OpenAI | $0.05 | $0.40 | $0.45 |
| GPT-5 Mini OpenAI | $0.25 | $2.00 | $2.25 |
| GPT-5 OpenAI | $1.25 | $10.00 | $11.25 |
| GPT OSS 20B OpenAI Price via Kilo Gateway | $0.02▼ 10% | $0.09▲ 200% | $0.11 |
| GPT OSS 120B OpenAI Price via Kilo Gateway | $0.03▲ 100% | $0.15▲ 163% | $0.18 |
| Claude Opus 4.1 Anthropic Price via Merge Gateway | $15.00▲ 11% | $75.00▲ 11% | $90.00 |
| Codestral 25.08 Mistral | $0.30 | $0.90 | $1.20 |
| Qwen3 30B-A3B Instruct 2507 Alibaba (Qwen) Price via Cortecs | $0.10 | $0.30 | $0.40 |
| Qwen Flash Alibaba (Qwen) | $0.05 | $0.40 | $0.45 |
| GLM-4.5-Air Z.ai (GLM) | $0.20 | $1.10 | $1.30 |
| Qwen3 Coder Flash Alibaba (Qwen) | $0.30 | $1.50 | $1.80 |
| GLM-4.5 Z.ai (GLM) | $0.60 | $2.20 | $2.80 |
| Qwen3 235B-A22B Thinking 2507 Alibaba (Qwen) Price via Venice AI | $0.45▲ 50% | $3.50▲ 17% | $3.95 |
| Qwen3 235B-A22B Instruct 2507 Alibaba (Qwen) Price via Cortecs | $0.20▲ 121% | $0.60▲ 3.1% | $0.80 |
| Voxtral Small Mistral | $0.10 | $0.40 | $0.50 |
| Kimi K2 Instruct Moonshot AI (Kimi) Price via NanoGPT | $0.40 | $1.80 | $2.20 |
| Kimi K2 0711 Moonshot AI (Kimi) Price via DevPass (LLM Gateway) | $0.57 | $2.30 | $2.87 |
| Gemini 2.5 Flash-Lite Google | $0.10 | $0.40 | $0.50 |
| Gemini 2.5 Flash-Lite Preview 06-17 Google Price via NanoGPT | $0.15 | $0.60 | $0.75 |
| Gemini 2.5 Flash Google | $0.30 | $2.50 | $2.80 |
| Gemini 2.5 Pro Google | $1.25 | $10.00 | $11.25 |
| o3-pro OpenAI | $20.00 | $80.00 | $100.00 |
| Gemini 2.5 Pro Preview 06-05 Google Price via Kilo Gateway | $0.63▼ 50% | $5.00▼ 50% | $5.63 |
| DeepSeek R1 0528 Qwen3 8B DeepSeek Price via Novita AI | $0.06 | $0.09 | $0.15 |
| DeepSeek R1 0528 DeepSeek Price via OpenRouter | $0.50 | $2.15 | $2.65 |
| Gemini 2.5 Flash Preview 05-20 Google Price via Jiekou.AI | $0.14 | $3.33 | $3.47 |
| Mistral Medium 3 Mistral Price via Vertex | $0.40 | $2.00 | $2.40 |
| Llama Prompt Guard 2 22MCheapest Meta Price via Helicone | $0.01 | $0.01 | $0.02 |
| Llama Prompt Guard 2 86M Meta Price via Helicone | $0.01 | $0.01 | $0.02 |
| Qwen3 30B A3B Alibaba (Qwen) Price via Helicone | $0.08 | $0.29 | $0.37 |
| o3 OpenAI | $2.00 | $8.00 | $10.00 |
| GPT-4.1 mini OpenAI | $0.40 | $1.60 | $2.00 |
| GPT-4.1 OpenAI | $2.00 | $8.00 | $10.00 |
| Pixtral Large (25.02) Mistral Price via Cortecs | $1.99 | $5.98 | $7.97 |
| Llama 4 Scout 17B Instruct Meta Price via DevPass (LLM Gateway) | $0.18 | $0.59 | $0.77 |
| Llama 4 Maverick 17B Instruct Meta Price via DevPass (LLM Gateway) | $0.27 | $0.85 | $1.12 |
| Qwen3-VL 30B-A3B Alibaba (Qwen) | $0.20 | $0.80 | $1.00 |
| DeepSeek V3 0324 DeepSeek Price via NanoGPT | $0.20 | $0.77 | $0.97 |
| Gemma 3 4B IT Google Price via Kilo Gateway | $0.04 | $0.05▼ 38% | $0.09 |
| Gemma 3 12B IT Google Price via Helicone | $0.05 | $0.10 | $0.15 |
| Gemma 3 27B IT Google Price via Kilo Gateway | $0.08 | $0.16 | $0.24 |
| DeepSeek-R1 DeepSeek Price via Merge Gateway | $1.35▲ 238% | $5.40▲ 218% | $6.75 |
| Qwen-MT Plus Alibaba (Qwen) | $2.46 | $7.37 | $9.83 |
| DeepSeek-V3 DeepSeek Price via Merge Gateway | $0.27▼ 5.9% | $1.10▼ 4.1% | $1.37 |
| Gemini 2.0 Flash-Lite Google Price via 302.AI | $0.07▲ 44% | $0.30▲ 43% | $0.38 |
| Llama-3.3-70B-Instruct Meta Price via Helicone | $0.13 | $0.39 | $0.52 |
| GPT-4o (2024-11-20) OpenAI | $2.50 | $10.00 | $12.50 |
| Mistral Large 2.1 Mistral Price via Helicone | $2.00 | $6.00 | $8.00 |
USD per 1M tokens, standard rates. Cached input is the price for tokens read from the prompt cache. DeepSeek prices are off-peak.
How to use this LLM pricing calculator
- Choose a provider, or leave it on All to compare every model.
- Enter how many input and output tokens you expect to use. The default is 1 million of each.
- Read the Total cost column. Click any column heading to sort by it.
The formula is: (input tokens × input price + output tokens × output price) ÷ 1,000,000. For example, 2M input and 500K output tokens on Claude Sonnet 5 cost (2 × $2.00) + (0.5 × $10.00) = $9.00.
LLM API pricing by provider
OpenAI API pricing
OpenAI offers about 46 models through its API. OpenAI released its newest model, GPT-6.1 Terra, on October 26, 2026. The lowest price belongs to GPT OSS 20B, at $0.09 per 1M output tokens.
On the other end, GPT-5.4 Pro is the most expensive model at $180.00 per 1M output tokens. The table below lists every OpenAI model with its input and output price. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| GPT-6.1 Terra | $1.00 | $5.00 | Oct 26, 2026 |
| GPT-6.1 Sol | $2.00 | $10.00 | Sep 29, 2026 |
| GPT-6 Luna | $0.10 | $0.50 | Sep 22, 2026 |
| GPT-6 Sol | $2.00 | $10.00 | Sep 22, 2026 |
| GPT-6 Astra | $10.00 | $50.00 | Sep 4, 2026 |
| GPT-6 Astra (Fast)Price via Vercel AI Gateway | $20.00 | $100.00 | Sep 4, 2026 |
| GPT-5.6 | $4.00 | $20.00 | Jul 9, 2026 |
| GPT-5.6 Luna | $0.20 | $1.20 | Jul 9, 2026 |
| GPT-5.6 SolPromo price, until at least Nov 21, 2026 | $4.00 | $20.00 | Jul 9, 2026 |
| GPT-5.6 Terra | $2.00 | $12.00 | Jul 9, 2026 |
| GPT-5.5 | $5.00 | $30.00 | Apr 23, 2026 |
| GPT-5.5 Pro | $30.00 | $180.00 | Apr 23, 2026 |
| GPT-5.4 mini | $0.75 | $4.50 | Mar 17, 2026 |
| GPT-5.4 nano | $0.20 | $1.25 | Mar 17, 2026 |
| GPT-5.4 | $2.50 | $15.00 | Mar 5, 2026 |
| GPT-5.4 Pro | $30.00 | $180.00 | Mar 5, 2026 |
| GPT-5.3 Chat (latest)Price via Merge Gateway | $1.75 | $14.00 | Mar 3, 2026 |
| GPT-5.3 Codex | $1.75 | $14.00 | Feb 5, 2026 |
| GPT-5.3 Codex Spark | $1.75 | $14.00 | Feb 5, 2026 |
| GPT-5.2 | $1.75 | $14.00 | Dec 11, 2025 |
| GPT-5.2 CodexPrice via Azure | $1.75▲ 1,150% | $14.00▲ 1,128% | Dec 11, 2025 |
| GPT-5.2 Pro | $21.00 | $168.00 | Dec 11, 2025 |
| GPT-5.1 | $1.25 | $10.00 | Nov 13, 2025 |
| GPT-5.1 CodexPrice via OpenCode | $1.07 | $8.50 | Nov 13, 2025 |
| GPT-5.1 Codex MaxPrice via Jiekou.AI | $1.19▲ 8.0% | $9.50▲ 5.6% | Nov 13, 2025 |
| GPT-5.1 Codex miniPrice via Jiekou.AI | $0.24▲ 8.0% | $1.90▲ 5.6% | Nov 13, 2025 |
| GPT-5 Pro | $15.00 | $120.00 | Oct 6, 2025 |
| GPT-5-CodexPrice via OpenCode | $1.07 | $8.50 | Sep 15, 2025 |
| GPT-5 | $1.25 | $10.00 | Aug 7, 2025 |
| GPT-5 Mini | $0.25 | $2.00 | Aug 7, 2025 |
| GPT-5 Nano | $0.05 | $0.40 | Aug 7, 2025 |
| GPT OSS 120BPrice via Kilo Gateway | $0.03▲ 100% | $0.15▲ 163% | Aug 5, 2025 |
| GPT OSS 20BPrice via Kilo Gateway | $0.02▼ 10% | $0.09▲ 200% | Aug 5, 2025 |
| o3-pro | $20.00 | $80.00 | Jun 10, 2025 |
| o3 | $2.00 | $8.00 | Apr 16, 2025 |
| GPT-4.1 | $2.00 | $8.00 | Apr 14, 2025 |
| GPT-4.1 mini | $0.40 | $1.60 | Apr 14, 2025 |
| GPT-4o (2024-11-20) | $2.50 | $10.00 | Nov 20, 2024 |
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| GPT Image 2.5 FlarePrice via Azure | $5.00 | $30.00 | Sep 8, 2026 |
| GPT Image 2.5 SunburstPrice via Azure | $5.00 | $30.00 | Sep 8, 2026 |
| GPT-5.4 Image 2Price via OpenRouter | $8.00 | $15.00 | Apr 21, 2026 |
| gpt-image-2 | $5.00 | $30.00 | Apr 21, 2026 |
| GPT-5 Image MiniPrice via OpenRouter | $2.50 | $2.00 | Oct 16, 2025 |
| GPT-5 ImagePrice via OpenRouter | $10.00 | $10.00 | Oct 14, 2025 |
| GPT-Image-1Price via Azure | $5.00 | $40.00 | Apr 24, 2025 |
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| GPT-Realtime-2.1 | $4.00 | $24.00 | Jul 6, 2026 |
- Cached input on the GPT-6, GPT-5.6, GPT-5.5, and GPT-5.4 families is billed at 10% of the normal input price.
- Batch and Flex processing cost 50% of standard rates. Fast mode (formerly Priority) costs 2x.
Claude API pricing
There are about 15 models from Anthropic in the price list below. Claude Fable 5.2 is the most recent addition, released on November 9, 2026. If you want the lowest cost, Claude Haiku 5.5 is the cheapest option right now at $0.50 per 1M output tokens.
However, Claude Opus 4.1 is the most expensive model at $75.00 per 1M output tokens. See all of them side by side in the table below. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Claude Fable 5.2 | $8.00 | $40.00 | Nov 9, 2026 |
| Claude Haiku 5.5 | $0.10 | $0.50 | Oct 7, 2026 |
| Claude Sonnet 5.5 | $2.00 | $10.00 | Sep 28, 2026 |
| Claude Opus 5.5 | $4.00 | $20.00 | Sep 22, 2026 |
| Claude Fable 5.1 | $10.00 | $50.00 | Sep 1, 2026 |
| Claude Opus 5 | $5.00 | $25.00 | Jul 24, 2026 |
| Claude Sonnet 5 | $2.00 | $10.00 | Jun 29, 2026 |
| Claude Fable 5 | $10.00 | $50.00 | Jun 7, 2026 |
| Claude Opus 4.8 | $5.00 | $25.00 | May 28, 2026 |
| Claude Opus 4.7 | $5.00 | $25.00 | Apr 14, 2026 |
| Claude Sonnet 4.6 | $3.00 | $15.00 | Feb 17, 2026 |
| Claude Opus 4.6 | $5.00 | $25.00 | Feb 4, 2026 |
| Claude Opus 4.5 | $5.00 | $25.00 | Nov 24, 2025 |
| Claude Haiku 4.5 | $1.00 | $5.00 | Oct 15, 2025 |
| Claude Opus 4.1Price via Merge Gateway | $15.00▲ 11% | $75.00▲ 11% | Aug 5, 2025 |
- Claude 4.6 and later models include the full 1M-token context window at standard prices, so a very long prompt costs the same per token as a short one.
- Cache hits cost 10% of the input price on most models, 5% on Opus 5.5, and 2.5% on Fable 5.1.
- Claude 4.7 and later use a newer tokenizer that produces around 30% more tokens for the same text.
- Choosing US-only processing adds a 1.1x multiplier on Claude 4.6 and later.
Gemini API pricing
The Google API has about 40 models you can pay for by the token. The newest one is Gemini 4 Carbon, launched on November 10, 2026. For the lowest bill, Gemma 4 26B A4B IT costs the least at $0.03 per 1M output tokens.
By contrast, Gemini 3.1 Pro Preview is the most expensive model at $12.00 per 1M output tokens. Every model and its price is in the table below. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Gemini 4 Carbon | $2.00 | $10.00 | Nov 10, 2026 |
| Gemini 4 Argon | $2.00 | $10.00 | Oct 13, 2026 |
| Gemini 3.8 FlashIntro price until Dec 31, 2026, then doubles | $0.75 | $3.75 | Sep 2, 2026 |
| Gemini 3.7 Flash | $0.75 | $3.75 | Aug 13, 2026 |
| Gemini 3.5 Flash Lite | $0.30 | $2.50 | Jul 21, 2026 |
| Gemini 3.6 Flash | $0.75 | $3.75 | Jul 21, 2026 |
| Gemma 4 12B ITPrice via NanoGPT | $0.05 | $0.25 | Jun 9, 2026 |
| Gemini 3.5 Flash | $1.50 | $9.00 | May 19, 2026 |
| Gemini 3.1 Flash Lite | $0.25 | $1.50 | May 7, 2026 |
| Gemma 4 26B A4B ITPrice via Kilo Gateway | $0.01▲ 10% | $0.03▲ 13% | Apr 2, 2026 |
| Gemma 4 31B ITPrice via DevPass (LLM Gateway) | $0.10▲ 33% | $0.25▲ 98% | Apr 2, 2026 |
| Gemini 3.1 Pro PreviewHigher price above 200K-token prompts | $2.00 | $12.00 | Feb 19, 2026 |
| Gemini 3 Flash Preview | $0.50 | $3.00 | Dec 17, 2025 |
| Gemini 3 Pro PreviewPrice via QiHang | $0.57 | $3.43 | Nov 18, 2025 |
| Gemini 2.5 Flash-Lite Preview 09-2025Price via 302.AI | $0.10 | $0.40 | Sep 25, 2025 |
| Gemini 2.5 Flash | $0.30 | $2.50 | Jun 17, 2025 |
| Gemini 2.5 Flash-Lite | $0.10 | $0.40 | Jun 17, 2025 |
| Gemini 2.5 Flash-Lite Preview 06-17Price via NanoGPT | $0.15 | $0.60 | Jun 17, 2025 |
| Gemini 2.5 Pro | $1.25 | $10.00 | Jun 17, 2025 |
| Gemini 2.5 Pro Preview 06-05Price via Kilo Gateway | $0.63▼ 50% | $5.00▼ 50% | Jun 5, 2025 |
| Gemini 2.5 Flash Preview 05-20Price via Jiekou.AI | $0.14 | $3.33 | May 20, 2025 |
| Gemma 3 12B ITPrice via Helicone | $0.05 | $0.10 | Mar 12, 2025 |
| Gemma 3 27B ITPrice via Kilo Gateway | $0.08 | $0.16 | Mar 12, 2025 |
| Gemma 3 4B ITPrice via Kilo Gateway | $0.04 | $0.05▼ 38% | Mar 12, 2025 |
| Gemini 2.0 Flash-LitePrice via 302.AI | $0.07▲ 44% | $0.30▲ 43% | Dec 11, 2024 |
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Nano Banana 2.1Price via Vercel AI Gateway | $1.50 | $7.50 | Oct 6, 2026 |
| Nano Banana 2 Lite | $0.25 | $30.00 | Jun 30, 2026 |
| Nano Banana 2 | $0.50 | $60.00 | May 28, 2026 |
| Nano Banana Pro | $2.00 | $120.00 | May 28, 2026 |
| Nano Banana 2 PreviewPrice via OpenRouter | $0.50 | $3.00▼ 95% | Feb 26, 2026 |
| Nano Banana Pro PreviewPrice via Kilo Gateway | $1.00▼ 50% | $6.00▼ 95% | Nov 20, 2025 |
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Gemini Omni 1.1 FlashPrice via OpenRouter | $1.50▼ 1.0% | $9.00▼ 1.0% | Aug 27, 2026 |
| Gemini Omni Flash Preview | $1.50 | $17.50 | Jun 30, 2026 |
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Gemini 3.5 Live Translate Preview | $3.50 | $21.00 | Jun 9, 2026 |
| Gemini 3.1 Flash TTS Preview | $1.00 | $20.00 | Apr 15, 2026 |
| Gemini 3.1 Flash Live Preview | $0.75 | $4.50 | Mar 26, 2026 |
| Gemini 2.5 Flash TTSPrice via Vertex | $0.50 | $10.00 | Sep 30, 2025 |
| Gemini 2.5 Pro TTSPrice via Vertex | $1.00 | $20.00 | Sep 30, 2025 |
| Gemini 2.5 Flash Preview TTS | $0.50 | $10.00 | May 1, 2025 |
| Gemini 2.5 Pro Preview TTS | $1.00 | $20.00 | May 1, 2025 |
- Gemini 3.1 Pro prompts over 200K tokens are billed at a higher rate.
DeepSeek API pricing
You can use about 18 models through the DeepSeek API. DeepSeek V4.1 Flash 1017 is the newest model in the family, which DeepSeek released on October 17, 2026. DeepSeek OCR 2 is the cheapest DeepSeek model at $0.03 per 1M output tokens.
At the top of the range, DeepSeek-R1 is the most expensive model at $5.40 per 1M output tokens. Use the table below to compare them all. Prices last checked on November 11, 2026.
DeepSeek charges different rates depending on the time of day. Peak hours are 01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday, and cost twice the off-peak rate. All other hours, including weekends, are off-peak.
| Model | Off-peak (in / out) | Peak (in / out) | Released |
|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | $0.15 / $0.60 | $0.30 / $1.20 | Oct 17, 2026 |
| DeepSeek V4.1 Flash | $0.03 / $0.12 | $0.06 / $0.24 | Sep 10, 2026 |
| DeepSeek V4 Pro | $0.66 / $1.98 | $1.32 / $3.96 | Aug 13, 2026 |
| DeepSeek V4 Pro 0813 | $0.66 / $1.98 | $1.32 / $3.96 | Aug 13, 2026 |
| DeepSeek V4 Flash 0731 | $0.04 / $0.09 | $0.09 / $0.18 | Jul 31, 2026 |
| DeepSeek V4 Flash 0423 | $0.14 / $0.28 | $0.28 / $0.56 | Apr 23, 2026 |
| DeepSeek V4 Pro 0423 | $1.65 / $3.30 | $3.30 / $6.60 | Apr 23, 2026 |
| DeepSeek OCR 2 | $0.03 / $0.03 | $0.06 / $0.06 | Jan 27, 2026 |
| DeepSeek Chat | $0.10 / $0.42 | $0.20 / $0.85 | Dec 1, 2025 |
| DeepSeek Reasoner | $0.24 / $0.73 | $0.48 / $1.45 | Dec 1, 2025 |
| DeepSeek V3.2 | $0.21 / $0.31 | $0.42 / $0.62 | Dec 1, 2025 |
| DeepSeek V3.1 | $0.25 / $0.95 | $0.50 / $1.90 | Aug 21, 2025 |
| DeepSeek-V3.1 | $0.19 / $0.71 | $0.38 / $1.42 | Aug 21, 2025 |
| DeepSeek R1 0528 Qwen3 8B | $0.06 / $0.09 | $0.12 / $0.18 | May 29, 2025 |
| DeepSeek R1 0528 | $0.50 / $2.15 | $1.00 / $4.30 | May 28, 2025 |
| DeepSeek V3 0324 | $0.20 / $0.77 | $0.40 / $1.54 | Mar 24, 2025 |
| DeepSeek-R1 | $1.35 / $5.40 | $2.70 / $10.80 | Jan 20, 2025 |
| DeepSeek-V3 | $0.27 / $1.10 | $0.54 / $2.20 | Dec 26, 2024 |
DeepSeek V4.1 Flash replaced V4 Flash on September 10, 2026 and uses the model name deepseek-flash. The calculator above shows off-peak prices, so double them if your requests run during peak hours.
Grok API pricing
The xAI API currently lists about 11 models. The latest release is Grok 4.7, out since September 21, 2026. On a tight budget, X-Ai/Grok-4-Fast-Reasoning costs the least at $0.50 per 1M output tokens.
Meanwhile, Grok 4.7 is the most expensive model at $6.00 per 1M output tokens. Compare every model side by side in the table below before you decide. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Grok 4.7Double price at 200K+ token prompts | $2.00 | $6.00 | Sep 21, 2026 |
| Grok 4.6 | $2.00 | $6.00 | Aug 12, 2026 |
| Grok 4.5 | $2.00 | $6.00 | Jul 8, 2026 |
| Grok 4.3 | $1.25 | $2.50 | Apr 17, 2026 |
| Grok Build 0.1 | $1.00 | $2.00 | Apr 16, 2026 |
| Grok 4.20 (Non-Reasoning) | $1.25 | $2.50 | Mar 9, 2026 |
| Grok 4.20 (Reasoning) | $1.25 | $2.50 | Mar 9, 2026 |
| Grok 4.20 Multi-Agent | $1.25 | $2.50 | Mar 9, 2026 |
| X-Ai/Grok-4-Fast-Non-ReasoningPrice via Helicone | $0.20▲ 11% | $0.50▲ 11% | Dec 18, 2025 |
| X-Ai/Grok-4-Fast-ReasoningPrice via Helicone | $0.20▲ 11% | $0.50▲ 11% | Dec 18, 2025 |
| x-AI/Grok-Code-Fast 1Price via Jiekou.AI | $0.19▲ 5.6% | $1.43▲ 5.6% | Sep 2, 2025 |
- Once a prompt reaches 200K tokens, xAI bills the whole request at double the standard rate.
GLM API pricing
Z.ai offers about 16 models through its API. Z.ai released its newest model, GLM-5.3-FlashX, on September 18, 2026. The lowest price belongs to GLM-4.7-FlashX, at $0.40 per 1M output tokens.
In comparison, GLM-5.2 is the most expensive model at $4.40 per 1M output tokens. The table below lists every Z.ai model with its input and output price. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| GLM-5.3-FlashX | $0.37 | $1.25 | Sep 18, 2026 |
| GLM-5.3-Flash | $0.15 | $0.50 | Aug 26, 2026 |
| GLM-5.3 | $1.40 | $4.40 | Aug 14, 2026 |
| GLM-5.2 | $1.40 | $4.40 | Jun 13, 2026 |
| GLM-5.1 | $1.40 | $4.40 | Apr 7, 2026 |
| GLM-5V-Turbo | $1.20 | $4.00 | Apr 1, 2026 |
| GLM-5-Turbo | $1.20 | $4.00 | Mar 16, 2026 |
| GLM-5 | $1.00 | $3.20 | Feb 12, 2026 |
| GLM-4.7-FlashPrice via Cortecs | $0.08▲ 33% | $0.48▲ 20% | Jan 19, 2026 |
| GLM-4.7-FlashX | $0.07 | $0.40 | Jan 19, 2026 |
| GLM-4.7 | $0.60 | $2.20 | Dec 22, 2025 |
| GLM-4.6V | $0.30 | $0.90 | Dec 8, 2025 |
| GLM-4.6 | $0.60 | $2.20 | Sep 30, 2025 |
| GLM-4.5V | $0.60 | $1.80 | Aug 11, 2025 |
| GLM-4.5 | $0.60 | $2.20 | Jul 28, 2025 |
| GLM-4.5-Air | $0.20 | $1.10 | Jul 28, 2025 |
Qwen API pricing
There are about 35 models from Alibaba in the price list below. Qwen 3.8 Max Prime is the most recent addition, released on September 23, 2026. If you want the lowest cost, Qwen3.5 2B is the cheapest option right now at $0.10 per 1M output tokens.
At the premium end, Qwen 3.8 Max Prime is the most expensive model at $12.00 per 1M output tokens. See all of them side by side in the table below. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Qwen 3.8 Max PrimePrice via Vercel AI Gateway | $4.00 | $12.00 | Sep 23, 2026 |
| Qwen3.8 Omni Flash | $0.15 | $0.47 | Sep 17, 2026 |
| Qwen3.8 Max 0902Price via Vercel AI Gateway | $2.00 | $6.00 | Sep 2, 2026 |
| Qwen3.8 Flash | $0.15 | $0.47 | Aug 26, 2026 |
| Qwen3.8 27BPrice via engy | $0.04▼ 55% | $0.32▼ 20% | Aug 14, 2026 |
| Qwen3.8 2.4T A95BPrice via DevPass (LLM Gateway) | $2.00 | $6.00 | Aug 12, 2026 |
| Qwen3.8 Max | $2.00 | $6.00 | Aug 3, 2026 |
| Qwen3.7 Flash | $0.03 | $0.13 | Jul 15, 2026 |
| Qwen3.7 Plus | $0.40 | $1.60 | Jun 2, 2026 |
| Qwen3.7 Max | $2.50 | $7.50 | May 21, 2026 |
| Qwen3.6 Flash | $0.25▲ 33% | $1.50▲ 33% | Apr 27, 2026 |
| Qwen3.6 27B | $0.60 | $3.60 | Apr 22, 2026 |
| Qwen3.6 35B-A3B | $0.38▲ 51% | $2.25▲ 52% | Apr 17, 2026 |
| Qwen3.6 Plus | $0.50 | $3.00 | Apr 2, 2026 |
| Qwen3.5 2BPrice via Requesty | $0.02 | $0.10 | Mar 2, 2026 |
| Qwen3.5 122B-A10B | $0.40 | $3.20 | Feb 23, 2026 |
| Qwen3.5 27B | $0.30 | $2.40 | Feb 23, 2026 |
| Qwen3.5 35B-A3B | $0.25 | $2.00 | Feb 23, 2026 |
| Qwen3.5 9BPrice via DevPass (LLM Gateway) | $0.10▲ 150% | $0.15 | Feb 23, 2026 |
| Qwen3.5 Flash | $0.10 | $0.40 | Feb 23, 2026 |
| Qwen3.5 Plus | $0.40 | $2.40 | Feb 16, 2026 |
| Qwen3.5 397B-A17B | $0.60 | $3.60 | Feb 15, 2026 |
| Qwen3 Coder NextPrice via Cortecs | $0.17▲ 55% | $0.89▲ 32% | Feb 3, 2026 |
| Qwen3 VL 30B A3B InstructPrice via DevPass (LLM Gateway) | $0.15 | $0.60 | Oct 4, 2025 |
| Qwen3 VL 235B A22B InstructPrice via DevPass (LLM Gateway) | $0.20 | $0.88 | Sep 23, 2025 |
| Qwen3-VL Plus | $0.20 | $1.60 | Sep 23, 2025 |
| Qwen Flash | $0.05 | $0.40 | Jul 28, 2025 |
| Qwen3 30B-A3B Instruct 2507Price via Cortecs | $0.10 | $0.30 | Jul 28, 2025 |
| Qwen3 Coder Flash | $0.30 | $1.50 | Jul 28, 2025 |
| Qwen3 235B-A22B Thinking 2507Price via Venice AI | $0.45▲ 50% | $3.50▲ 17% | Jul 25, 2025 |
| Qwen3 235B-A22B Instruct 2507Price via Cortecs | $0.20▲ 121% | $0.60▲ 3.1% | Jul 21, 2025 |
| Qwen3 30B A3BPrice via Helicone | $0.08 | $0.29 | Apr 28, 2025 |
| Qwen3-VL 30B-A3B | $0.20 | $0.80 | Apr 2025 |
| Qwen-MT Plus | $2.46 | $7.37 | Jan 2025 |
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Qwen3-Omni Flash | $0.43 | $1.66 | Sep 15, 2025 |
Mistral API pricing
The Mistral API has about 13 models you can pay for by the token. The newest one is Mistral Large 4, launched on October 6, 2026. For the lowest bill, Ministral 3 3B costs the least at $0.10 per 1M output tokens.
On the pricier side, Mistral Medium 3.5 is the most expensive model at $7.50 per 1M output tokens. Every model and its price is in the table below. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Mistral Large 4 | $0.70▲ 2.9% | $2.10▲ 0.5% | Oct 6, 2026 |
| Mistral Large 4.0.8 | $0.70 | $2.10 | Oct 6, 2026 |
| Mistral Medium 3.5 | $1.50 | $7.50 | Apr 29, 2026 |
| Mistral Small 4 | $0.15 | $0.60 | Mar 16, 2026 |
| Ministral 14BPrice via Vercel AI Gateway | $0.20 | $0.20 | Dec 2, 2025 |
| Ministral 3 14B | $0.20 | $0.20 | Dec 2, 2025 |
| Ministral 3 3B | $0.10 | $0.10 | Dec 2, 2025 |
| Ministral 3 8B | $0.15 | $0.15 | Dec 2, 2025 |
| Codestral 25.08 | $0.30 | $0.90 | Jul 30, 2025 |
| Voxtral Small | $0.10 | $0.40 | Jul 15, 2025 |
| Mistral Medium 3Price via Vertex | $0.40 | $2.00 | May 7, 2025 |
| Pixtral Large (25.02)Price via Cortecs | $1.99 | $5.98 | Apr 8, 2025 |
| Mistral Large 2.1Price via Helicone | $2.00 | $6.00 | Nov 18, 2024 |
Meta Llama API pricing
You can use about 14 models through the Meta API. Muse Ember is the newest model in the family, which Meta released on November 9, 2026. Llama Prompt Guard 2 22M is the cheapest Meta model at $0.01 per 1M output tokens.
At the high end, Muse Wonder is the most expensive model at $20.00 per 1M output tokens. Use the table below to compare them all. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Muse Ember | $0.10 | $0.50 | Nov 9, 2026 |
| Muse Spark OpenPrice via Ollama Cloud | $1.25 | $4.25 | Oct 15, 2026 |
| Muse Wonder | $4.00 | $20.00 | Oct 12, 2026 |
| Muse Spark 1.3 | $1.25 | $4.25 | Sep 2, 2026 |
| Muse Spark 1.3 Contributor | $0.10 | $0.20 | Sep 2, 2026 |
| Muse Glimmer 30BPrice via EmpirioLabs AI | $0.20 | $0.80 | Aug 10, 2026 |
| Muse Spark 1.2 | $1.25 | $4.25 | Aug 5, 2026 |
| Muse Spark 1.2 Contributor | $0.10 | $0.20 | Aug 5, 2026 |
| Muse Spark 1.1 | $1.25 | $4.25 | Apr 8, 2026 |
| Llama Prompt Guard 2 22MPrice via Helicone | $0.01 | $0.01 | Apr 29, 2025 |
| Llama Prompt Guard 2 86MPrice via Helicone | $0.01 | $0.01 | Apr 29, 2025 |
| Llama 4 Maverick 17B InstructPrice via DevPass (LLM Gateway) | $0.27 | $0.85 | Apr 5, 2025 |
| Llama 4 Scout 17B InstructPrice via DevPass (LLM Gateway) | $0.18 | $0.59 | Apr 5, 2025 |
| Llama-3.3-70B-InstructPrice via Helicone | $0.13 | $0.39 | Dec 6, 2024 |
Kimi API pricing
The Moonshot AI API currently lists about 12 models. The latest release is Kimi K3.1 Pro Highspeed, out since October 23, 2026. On a tight budget, Kimi K2 Instruct costs the least at $1.80 per 1M output tokens.
On the other end, Kimi K3.1 Pro Highspeed is the most expensive model at $20.00 per 1M output tokens. Compare every model side by side in the table below before you decide. Prices last checked on November 11, 2026.
| Model | Input / 1M | Output / 1M | Released |
|---|---|---|---|
| Kimi K3.1 Pro | $2.00 | $10.00 | Oct 23, 2026 |
| Kimi K3.1 Pro Highspeed | $4.00 | $20.00 | Oct 23, 2026 |
| Kimi K2.8 PreviewPrice via DevPass (LLM Gateway) | $1.00 | $4.00 | Sep 11, 2026 |
| Kimi K3 | $3.00 | $15.00 | Jul 16, 2026 |
| Kimi K2.7 Code | $0.95 | $4.00 | Jun 12, 2026 |
| Kimi K2.7 Code HighSpeed | $1.90 | $8.00 | Jun 12, 2026 |
| Kimi K2.6 | $0.95 | $4.00 | Apr 20, 2026 |
| Kimi K2.5Price via NanoGPT | $0.38 | $2.00 | Jan 27, 2026 |
| Kimi K2 ThinkingPrice via Helicone | $0.48 | $2.00 | Nov 6, 2025 |
| Kimi K2 Instruct 0905Price via Helicone | $0.50 | $2.00 | Sep 5, 2025 |
| Kimi K2 0711Price via DevPass (LLM Gateway) | $0.57 | $2.30 | Jul 11, 2025 |
| Kimi K2 InstructPrice via NanoGPT | $0.40 | $1.80 | Jul 11, 2025 |
How LLM API pricing works
Input tokens vs output tokens
Input tokens are everything you send: your prompt, instructions, and any documents. Output tokens are what the model writes back. Almost every model charges more for output, from 1.0x on Llama Prompt Guard 2 22M to 23.3x on Gemini 2.5 Flash Preview 05-20. Reasoning models also bill their hidden thinking tokens as output, which can make a short visible answer cost more than expected.
Prompt caching discounts
If you send the same long instructions or documents again and again, providers can cache them and charge much less when they're reused.
| Provider | Cached input price |
|---|---|
| OpenAI (GPT-6, 5.6, 5.5, 5.4) | 10% of input price |
| Anthropic | 10% on most models, 5% on Opus 5.5, 2.5% on Fable 5.1 |
| Google Gemini 3.1 Pro | $0.20 per 1M |
| xAI Grok 4.7 | $0.50 per 1M |
| DeepSeek V4.1 Flash | $0.003 per 1M off-peak |
Anthropic also charges to write to the cache: 1.25x the input price for a 5-minute cache or 2x for a 1-hour cache.
Batch processing discounts
| Provider | Batch discount |
|---|---|
| OpenAI | 50% (Batch and Flex) |
| Anthropic | 50% on input and output |
| Google Gemini | 50% |
| xAI | 20% on Grok 4.3 and Grok 4.20 only |
Long-context surcharges
- Gemini 3.1 Pro: prompts over 200K tokens are billed at a higher rate.
- Grok: prompts of 200K tokens or more are billed at double the rate for the whole request.
- OpenAI: prompts over 272K input tokens have a separate long-prompt surcharge.
- Claude 4.6 and later: no surcharge across the 1M context window.
Frequently asked questions
Which LLM API is the cheapest right now?
Among the models in this calculator, Llama Prompt Guard 2 22M (Meta) has the lowest combined price at $0.01 / $0.01 per 1M input and output tokens. The lowest input price, $0.01 per 1M tokens, belongs to Llama Prompt Guard 2 22M and Llama Prompt Guard 2 86M.
Is ChatGPT API pricing the same as OpenAI API pricing?
Yes. There is no separate ChatGPT API. Developers call GPT models through the OpenAI API and pay per token. ChatGPT subscriptions like Plus are a different product and don't include API credits.
How much does 1 million tokens cost?
It depends on the model and on whether the tokens are input or output. For 1M input plus 1M output tokens, costs range from $0.02 on Llama Prompt Guard 2 22M to $210.00 on GPT-5.5 Pro. Enter your own numbers in the calculator above to compare.
Why are output tokens more expensive than input tokens?
Almost every model in this table charges more for output. The gap ranges from 1.0x on Llama Prompt Guard 2 22M to 23.3x on Gemini 2.5 Flash Preview 05-20, so tasks that generate long answers cost more than tasks that read long documents.
Do Claude models count tokens the same way as GPT models?
No. Anthropic says Claude 4.7 and later models use a newer tokenizer that produces about 30% more tokens for the same text. The same prompt can therefore cost more on newer Claude models than the price per token suggests.
Price data from Models.dev, checked against OpenAI, Anthropic, Google, DeepSeek, Z.ai (GLM), Alibaba (Qwen), Mistral, Meta, Moonshot AI (Kimi), and xAI's pricing pages. Prices can change at any time, so confirm on the provider's page before committing to a budget.
