Model Pricing Calculator
What would the same work cost on a different model? Enter token counts, or drop in a Claude Code or Codex session to price its real usage on every model below.
Token counts
Shorthand works too, such as 250k or 1.5M.
Billed at the full input price.
Read from a prompt cache.
Includes reasoning tokens.
Cache writes
Anthropic charges extra to write to its prompt cache: 1.25 times the input price for an entry that lasts five minutes, and twice the input price for one that lasts an hour. Other providers bill these tokens as uncached input.
Or price a real session
The session's token counts fill in the fields, and the table prices them on every model.
Drop Claude Code or Codex session files here
Files are read in your browser. Nothing is uploaded.
- Claude Code:
~/.claude/. Subagents live in the folder with the same name, so drop both to include them.projects/ <project>/ <session>.jsonl - Codex:
~/.codex/.sessions/ <year>/ <month>/ <day>/ - Both folders are hidden. In the macOS file picker, press ⌘⇧. to show them.
Comparison
Prices are US dollars per million tokens, last updated October 4, 2026. The cost column prices the token counts above.
| Cache writes 5 minutes / 1 hour | ||||||
|---|---|---|---|---|---|---|
| GPT-6 Astra OpenAI | $10.00 | $1.00 | Billed as uncached input | $50.00 | $60.00 | $60.00 |
| Claude Fable 5.1 Anthropic | $10.00 | $0.25 | $12.50 / $20.00 | $50.00 | $60.00 | $60.00 |
| Claude Opus 5.5 Anthropic | $4.00 | $0.20 | $5.00 / $8.00 | $20.00 | $24.00 | $24.00 |
| Kimi K3 Moonshot AI | $3.00 | $0.30 | Billed as uncached input | $15.00 | $18.00 | $18.00 |
| Gemini 3.1 Pro Preview ≤200K Google | $2.00 | $0.20 | Billed as uncached input | $12.00 | $14.00 | $14.00 |
| GPT-6.1 Sol OpenAI | $2.00 | $0.10 | Billed as uncached input | $10.00 | $12.00 | $12.00 |
| Claude Sonnet 5.5 Anthropic | $2.00 | $0.20 | $2.50 / $4.00 | $10.00 | $12.00 | $12.00 |
| Grok 4.7 <200K xAI | $2.00 | $0.50 | Billed as uncached input | $6.00 | $8.00 | $8.00 |
| Qwen3.8-Max Alibaba Cloud | $1.65 | $0.206 | Billed as uncached input | $4.951 | $6.601 | $6.60 |
| Claude Haiku 4.5 Anthropic | $1.00 | $0.10 | $1.25 / $2.00 | $5.00 | $6.00 | $6.00 |
| GLM-5.3 Z.ai | $1.40 | $0.26 | Billed as uncached input | $4.40 | $5.80 | $5.80 |
| DeepSeek V4 Pro peak DeepSeek | $1.32 | $0.044 | Billed as uncached input | $3.96 | $5.28 | $5.28 |
| Kimi K2.7 Code Moonshot AI | $0.95 | $0.19 | Billed as uncached input | $4.00 | $4.95 | $4.95 |
| Gemini 3.8 Flash Google | $0.75 | $0.075 | Billed as uncached input | $3.75 | $4.50 | $4.50 |
| Gemini 3.5 Flash-Lite Google | $0.30 | $0.03 | Billed as uncached input | $2.50 | $2.80 | $2.80 |
| DeepSeek V4 Pro off-peak DeepSeek | $0.66 | $0.022 | Billed as uncached input | $1.98 | $2.64 | $2.64 |
| DeepSeek V4.1 Flash peak DeepSeek | $0.30 | $0.006 | Billed as uncached input | $1.20 | $1.50 | $1.50 |
| DeepSeek V4.1 Flash off-peak DeepSeek | $0.15 | $0.003 | Billed as uncached input | $0.60 | $0.75 | $0.75 |
| GLM-5.3-Flash Z.ai | $0.15 | $0.03 | Billed as uncached input | $0.50 | $0.65 | $0.65 |
| GPT-6 Luna OpenAI | $0.10 | $0.01 | Billed as uncached input | $0.50 | $0.60 | $0.60 |
| Qwen3.8-Flash Alibaba Cloud | $0.113 | $0.014 | Billed as uncached input | $0.382 | $0.495 | $0.50 |
- Pricing a session on another model assumes that model would use the same tokens. Models tokenize text differently and write different amounts of output and reasoning, so treat other models' costs as estimates.
- Codex sessions don't report cache writes, so the Claude prices for a Codex session leave out Anthropic's cache-write premium.
- Prices marked ≤200K or <200K only cover prompts up to 200,000 tokens.
- Tool fees, such as web searches, aren't included.