Claude Sonnet API pricing · Updated
Claude Sonnet 5.5, the current Sonnet model, costs $2.00 per million input tokens and $10.00 per million output tokens on the Claude API. Cached input costs $0.10 per million and the Batch API halves both prices. Sonnet 5 costs the same $2.00 and $10.00; older Sonnet 4.6 and 4.5 cost $3.00 and $15.00. Only Sonnet 4.5 is being retired, on November 30, 2026.
| Model | Input | Cache read | 5-min cache write | Output | Batch in / out | Status |
|---|---|---|---|---|---|---|
Claude Sonnet 5.5claude-sonnet-5-5 | $2.00 | $0.10 | $2.50 | $10.00 | $1.00 / $5.00 | Current |
Claude Sonnet 5claude-sonnet-5 | $2.00 | $0.20 | $2.50 | $10.00 | $1.00 / $5.00 | Available |
Claude Sonnet 4.6claude-sonnet-4-6 | $3.00 | $0.30 | $3.75 | $15.00 | $1.50 / $7.50 | Available |
Claude Sonnet 4.5claude-sonnet-4-5 | $3.00 | $0.30 | $3.75 | $15.00 | $1.50 / $7.50 | Deprecated, shuts down November 30, 2026 |
Claude Sonnet 4claude-sonnet-4 | $3.00 | $0.30 | $3.75 | $15.00 | $1.50 / $7.50 | Shut down June 15, 2026 |
Prices from Anthropic’s pricing page and deprecation schedule, for Anthropic’s own API. Ganivra uses these same rates to price Claude Sonnet calls. Upcoming changes across providers are on the price change and shutdown tracker.
Worked out from the prices above, at standard rates with no Batch discount.
| Request | Claude Sonnet 5.5 | Claude Sonnet 4.6 | Per 1,000 on Claude Sonnet 5.5 |
|---|---|---|---|
| Chat reply3,000 input tokens, 500 output | $0.011 | $0.017 | $11.00 |
| Document summary30,000 input tokens, 1,500 output | $0.075 | $0.113 | $75.00 |
| Agent task80,000 input tokens (60,000 cached), 6,000 output | $0.106 | $0.168 | $106 |
The same agent task (80,000 input tokens (60,000 cached), 6,000 output) on other models, per 1,000 requests. Quality differs between models, so test on your own prompts before switching.
| Model | Input / output per 1M | Agent task per 1,000 | vs Claude Sonnet 5.5 |
|---|---|---|---|
| Claude Haiku 5.5 | $0.10 / $0.50 | $5.60 | −95% |
| DeepSeek V4.1 Flash on Together AI | $0.30 / $1.20 | $13.56 | −87% |
| Gemini 3.8 Flash | $0.75 / $3.75 | $42.00 | −60% |
| Claude Sonnet 5.5 | $2.00 / $10.00 | $106 | — |
| GPT-6.1 Sol | $2.00 / $10.00 | $106 | Same price |
| Gemini 4 Argon | $2.00 / $10.00 | $106 | Same price |
| Claude Opus 5.5 | $4.00 / $20.00 | $212 | +100% |
To see what your own calls would cost on each of these, Ganivra reprices your recorded usage in model comparison.
Head-to-head price comparisons: Claude Sonnet 5.5 vs GPT-6.1 Sol · Claude Sonnet 5.5 vs Gemini 4 Argon · Claude Sonnet 5.5 vs GPT-6.1 Terra · Claude Sonnet 5.5 vs DeepSeek V4.1 Flash · Claude Opus 5.5 vs Claude Sonnet 5.5 · Claude Sonnet 5.5 vs Claude Haiku 5.5 · all comparisons.
Ganivra prices every Claude Sonnet call at these rates and adds it up by customer, feature and workflow, next to what each customer pays. You see who is profitable on Claude Sonnet and who isn’t.
Claude Sonnet 5.5 costs $2.00 per million input tokens and $10.00 per million output tokens. A typical chat reply of 3,000 input and 500 output tokens costs about 1 cent. Cached input is $0.10 per million, and the Batch API halves the price.
Only Sonnet 4.5. It retires on the Claude API on November 30, 2026, and Anthropic recommends moving to Sonnet 5.5. Sonnet 4 already shut down on June 15, 2026. Sonnet 5.5, Sonnet 5 and Sonnet 4.6 are active, and Sonnet 5.5 will not be retired before September 28, 2027.
Per token, yes: $2.00 input and $10.00 output against $3.00 and $15.00, 33% less. But Sonnet 5.5 uses a newer tokenizer that counts about 30% more tokens for the same text, so the same prompt costs roughly 13% less, not 33%.
Input and output cost the same: $2.00 and $10.00 per million tokens. The difference is caching: cache reads cost $0.10 per million on Sonnet 5.5 and $0.20 on Sonnet 5, so cache-heavy workloads such as agents are cheaper on 5.5.
No. Sonnet 5 launched at $2.00 input and $10.00 output as introductory pricing, with a planned rise to $3.00 and $15.00 on September 1, 2026. Anthropic cancelled the rise, and $2.00 and $10.00 is now the standard price.
Claude Opus 5.5 costs $4.00 input and $20.00 output per million tokens, 2 times the price of Sonnet 5.5. Claude Haiku 5.5 costs $0.10 and $0.50 for prompts up to 100,000 tokens, 20 times cheaper than Sonnet.
Reading from the cache costs $0.10 per million tokens on Sonnet 5.5, 5% of the normal input price. Writing to the cache costs $2.50 per million for 5 minutes or $4.00 for 1 hour. Caching pays off after the first cache read.
No. Sonnet 4.6 and later include the full 1M-token context window at the standard price, so a 900,000-token request costs the same per token as a short one.
Tag each Sonnet call with the customer it served, then price it at these rates and add it up by customer. Ganivra does this automatically with two lines of the Python SDK, and shows each customer's Sonnet cost next to what they pay you.
Looking for another model? See all 165 supported models and prices.