LLM API prices · 71 models · checked Oct 23, 2026
LLM API prices, and the cheapest model for your job
Every price is copied from the provider's own pricing page, with the date. Tell the picker your job and volume and it does the math on all 29 current models.
Short answerThe cheapest current model for a 1,500-token chatbot reply is Claude Haiku 5.5 (tied with GPT-6 Luna, same list price) at $0.35 per 1,000 replies; the most expensive is GPT-5.4 Pro at $117, a 334x spread for the same request. The median current model costs $7.00. 1 million chatbot replies on the cheapest: $350.
Cheapest by job
Each card opens the full ranking for that job.
All models
Prices in USD per 1M tokens; the last column is 1,000 chatbot replies (1,500 tokens in, 400 out). Click a column to sort. Context window from each provider's model page.
29 models shown
| GPT-5 nanodeprecated | OpenAI | $0.05 | $0.4 | $0.005 | 400K | $0.23 |
| GPT-4.1 nanoshut down | OpenAI | $0.1 | $0.4 | $0.025 | 1.05M | $0.31 |
| Gemini 2.5 Flash-Litelimited access | $0.1 | $0.4 | $0.01 | 1.05M | $0.31 | |
| Claude Haiku 5.5 | Anthropic | $0.1 | $0.5 | $0.01 | 1M | $0.35 |
| GPT-6 Luna | OpenAI | $0.1 | $0.5 | $0.01 | 1.05M | $0.35 |
| GPT-4o minideprecated | OpenAI | $0.15 | $0.6 | $0.075 | 128K | $0.47 |
| GPT-5.6 Luna | OpenAI | $0.2 | $1.2 | $0.02 | 1.05M | $0.78 |
| GPT-5.4 nanodeprecated | OpenAI | $0.2 | $1.25 | $0.02 | 400K | $0.80 |
| DeepSeek Flash | DeepSeek | $0.3 | $1.2 | $0.006 | 1M | $0.93 |
| Gemini 3.1 Flash-Lite | $0.25 | $1.5 | $0.025 | 1.05M | $0.97 | |
| GPT-5 minideprecated | OpenAI | $0.25 | $2 | $0.025 | 400K | $1.18 |
| GPT-4.1 minideprecated | OpenAI | $0.4 | $1.6 | $0.1 | 1.05M | $1.24 |
| GPT-3.5 Turboshut down | OpenAI | $0.5 | $1.5 | n/a | 16.4K | $1.35 |
| GPT-3.5 Turbo (0125)no spec page | OpenAI | $0.5 | $1.5 | n/a | n/a | $1.35 |
| Gemini 3.5 Flash-Lite | $0.3 | $2.5 | $0.03 | 1.05M | $1.45 | |
| Gemini 2.5 Flashlimited access | $0.3 | $2.5 | $0.03 | 1.05M | $1.45 | |
| Gemini 3 Flash Previewpreview | $0.5 | $3 | $0.05 | 1.05M | $1.95 | |
| Gemini 3.8 Flash | $0.75 | $3.75 | $0.075 | 1.05M | $2.62 | |
| Gemini 3.6 Flash | $0.75 | $3.75 | $0.075 | 1.05M | $2.62 | |
| Claude Haiku 3.5retired | Anthropic | $0.8 | $4 | $0.08 | n/a | $2.80 |
| GPT-5.4 mini | OpenAI | $0.75 | $4.5 | $0.075 | 400K | $2.93 |
| o4-minishut down | OpenAI | $1.1 | $4.4 | $0.275 | 200K | $3.41 |
| o3-minishut down | OpenAI | $1.1 | $4.4 | $0.55 | 200K | $3.41 |
| Claude Haiku 4.5legacy | Anthropic | $1 | $5 | $0.1 | 200K | $3.50 |
| DeepSeek V4 Pro | DeepSeek | $1.32 | $3.96 | $0.044 | 1M | $3.56 |
| GPT-5.1deprecated | OpenAI | $1.25 | $10 | $0.125 | 400K | $5.88 |
| GPT-5deprecated | OpenAI | $1.25 | $10 | $0.125 | 400K | $5.88 |
| GPT-5 Search APIno spec page | OpenAI | $1.25 | $10 | $0.125 | n/a | $5.88 |
| Gemini 2.5 Prolimited access | $1.25 | $10 | $0.125 | 1.05M | $5.88 | |
| GPT-4.1deprecated | OpenAI | $2 | $8 | $0.5 | 1.05M | $6.20 |
| o3deprecated | OpenAI | $2 | $8 | $0.5 | 200K | $6.20 |
| Claude Sonnet 5.5 | Anthropic | $2 | $10 | $0.1 | 1M | $7.00 |
| Claude Sonnet 5legacy | Anthropic | $2 | $10 | $0.2 | 1M | $7.00 |
| GPT-6.1 Sol | OpenAI | $2 | $10 | $0.1 | 1.05M | $7.00 |
| GPT-6 Sol | OpenAI | $2 | $10 | $0.2 | 1.05M | $7.00 |
| Gemini 4 Argon | $2 | $10 | $0.1 | 1.05M | $7.00 | |
| GPT-4odeprecated | OpenAI | $2.5 | $10 | $1.25 | 128K | $7.75 |
| GPT-5.6 Terra | OpenAI | $2 | $12 | $0.2 | 1.05M | $7.80 |
| Gemini 3.1 Pro Previewpreview | $2 | $12 | $0.2 | 1.05M | $7.80 | |
| GPT-5.2 | OpenAI | $1.75 | $14 | $0.175 | 400K | $8.22 |
| GPT-5.3 Codexdeprecated | OpenAI | $1.75 | $14 | $0.175 | 400K | $8.22 |
| GPT-5.4 | OpenAI | $2.5 | $15 | $0.25 | 1.05M | $9.75 |
| Claude Sonnet 4.6legacy | Anthropic | $3 | $15 | $0.3 | 1M | $10.50 |
| Claude Sonnet 4.5deprecated | Anthropic | $3 | $15 | $0.3 | 200K | $10.50 |
| Claude Sonnet 4retired | Anthropic | $3 | $15 | $0.3 | n/a | $10.50 |
| GPT-4o (2024-05-13)no spec page | OpenAI | $5 | $15 | n/a | n/a | $13.50 |
| Claude Opus 5.5 | Anthropic | $4 | $20 | $0.2 | 1M | $14.00 |
| GPT-5.6 Sol | OpenAI | $4 | $20 | $0.4 | 1.05M | $14.00 |
| Claude Opus 5legacy | Anthropic | $5 | $25 | $0.5 | 1M | $17.50 |
| Claude Opus 4.8legacy | Anthropic | $5 | $25 | $0.5 | 1M | $17.50 |
| Claude Opus 4.7legacy | Anthropic | $5 | $25 | $0.5 | 1M | $17.50 |
| Claude Opus 4.6legacy | Anthropic | $5 | $25 | $0.5 | 1M | $17.50 |
| Claude Opus 4.5legacy | Anthropic | $5 | $25 | $0.5 | 200K | $17.50 |
| GPT-5.5 | OpenAI | $5 | $30 | $0.5 | 1.05M | $19.50 |
| chat-latest | OpenAI | $5 | $30 | $0.5 | 400K | $19.50 |
| GPT-4 Turbo (2024-04-09)no spec page | OpenAI | $10 | $30 | n/a | n/a | $27.00 |
| Claude Fable 5.1 | Anthropic | $10 | $50 | $0.25 | 1M | $35.00 |
| Claude Fable 5legacy | Anthropic | $10 | $50 | $1 | 1M | $35.00 |
| GPT-6 Astra | OpenAI | $10 | $50 | $1 | 1.05M | $35.00 |
| o1shut down | OpenAI | $15 | $60 | $7.5 | 200K | $46.50 |
| GPT-5.6 Cyber | OpenAI | $12.5 | $75 | $1.25 | 400K | $48.75 |
| GPT-5.5 Cyberno spec page | OpenAI | $12.5 | $75 | $1.25 | n/a | $48.75 |
| Claude Opus 4.1retired | Anthropic | $15 | $75 | $1.5 | n/a | $52.50 |
| Claude Opus 4retired | Anthropic | $15 | $75 | $1.5 | n/a | $52.50 |
| o3-prodeprecated | OpenAI | $20 | $80 | n/a | 200K | $62.00 |
| GPT-4 (0613)no spec page | OpenAI | $30 | $60 | n/a | n/a | $69.00 |
| GPT-5 Prodeprecated | OpenAI | $15 | $120 | n/a | 400K | $70.50 |
| GPT-5.2 Pro | OpenAI | $21 | $168 | n/a | 400K | $98.70 |
| GPT-5.5 Pro | OpenAI | $30 | $180 | n/a | 1.05M | $117 |
| GPT-5.4 Pro | OpenAI | $30 | $180 | n/a | 1.05M | $117 |
| o1-proshut down | OpenAI | $150 | $600 | n/a | 200K | $465 |
Newest
Your first visit here, so these are the newest.
- o4-mini added to the table: $1.1 input / $4.4 output per 1M tokens.price change Oct 23
- o3-pro added to the table: $20 input / $80 output per 1M tokens.price change Oct 23
- o3-mini added to the table: $1.1 input / $4.4 output per 1M tokens.price change Oct 23
- o3 added to the table: $2 input / $8 output per 1M tokens.price change Oct 23
- o1-pro added to the table: $150 input / $600 output per 1M tokens.price change Oct 23
- o1 added to the table: $15 input / $60 output per 1M tokens.price change Oct 23
- GPT-6 Sol added to the table: $2 input / $10 output per 1M tokens.price change Oct 23
- GPT-5.6 Sol added to the table: $4 input / $20 output per 1M tokens.price change Oct 23
- GPT-5.6 Cyber added to the table: $12.5 input / $75 output per 1M tokens.price change Oct 23
- GPT-5.5 Pro added to the table: $30 input / $180 output per 1M tokens.price change Oct 23
- GPT-5.5 Cyber added to the table: $12.5 input / $75 output per 1M tokens.price change Oct 23
- GPT-5.4 Pro added to the table: $30 input / $180 output per 1M tokens.price change Oct 23
Also most read on the site
- crates.io download count: 80% of ours arrived with a new versionArticle · 420 views
- Tim Sweeney told Kenney "up yours" on 2.5M views. Kenney is rightArticle · 45 views
- App Store review time: 64 real decisions, median 73 hoursArticle · 44 views
- We cut the AI replies. Daily impressions fell from 447,575 to 5,686Article · 22 views
- Run an LLM in the browser: WebGPU did 63 tokens/s, WASM did 1.6Article · 20 views
Views: Vercel Web Analytics, Oct 17 to Oct 24 (UTC). "New since your last visit" is remembered in your browser only.
Head-to-head comparisons
Only pairs people actually search for (checked with Google autocomplete).
Method
Prices: each provider's pricing page, copied into a dated snapshot (2 checks so far; latest Oct 23, 2026). Current = not marked deprecated, legacy, shut down or retired on the provider's own model page, and context windows and input types come from those pages too (checked Oct 24, 2026). The picker bills each job the cheapest way a price list allows (cache reads for repeated context, batch for work that can wait) and leaves out cache-write charges. It ranks cost, not quality. Cutting the bill further: stop paying twice for the same answer, and retry without burning tokens.
Sources
- Anthropic: https://platform.claude.com/docs/en/about-claude/pricing
- OpenAI: https://developers.openai.com/api/docs/pricing
- Google: https://ai.google.dev/gemini-api/docs/pricing
- DeepSeek: https://api-docs.deepseek.com/quick_start/pricing
- Table and update script: github.com/Mattbusel/llm-prices (MIT).