Balanced Claude model for coding, analysis, agent workflows, and cost controlClaude Sonnet 4.5 (latest) by anthropic is offered by 35 providers on this page. Public prices are shown for 33 of them. Its official list price is $3.00 per 1M input tokens and $15.00 per 1M output tokens. Prices vary about 1.8× across the providers that host it.
Median output speed is 49 tokens per second, with 13.63s to the first token.
The context window is 200,000 tokens at the reference host, but hosts report different limits, from 200,000 to 1,000,000, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and PDF. Its training knowledge cuts off at 2025-07-31.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| Poe anthropic/claude-sonnet-4.5 | Gateway | $2.58 | $12.88 | $0.26 | $3.22 | 983,040 ⚠ | 32,768 | |
| AnthropicOfficial | First-party | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 | deprecated |
| Perplexity Agent | Gateway | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 | |
| LLM Gateway | Gateway | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 | |
| LLM Gateway | Gateway | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 | |
| LLM Gateway | Gateway | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 8,192 | |
| Requesty | Gateway | $3.00 tiered >200K: $6.00 | $15.00 | $0.30 | $3.75 | 1,000,000 ⚠ | 64,000 | |
| DigitalOcean anthropic-claude-4.5-sonnet | Cloud | $3.00 tiered >200K: $6.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 | |
| NanoGPT anthropic/claude-sonnet-4.5 | Gateway | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 | |
| NanoGPT anthropic/claude-sonnet-4.5:thinking | Gateway | $3.00 | $15.00 | $0.30 | $3.75 | 200,000 | 64,000 |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Interleaved thinking (reasoning between tool calls) is declared by 4 of 35 providers.
3 / 35 providers expose no reasoning control (reasoning_options: []).
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Pro | dataset: public | 43.6 | resolve rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.