Claude Haiku 4.5 vs GPT-5 vs Claude Sonnet 4.5
Pick two to four models to compare benchmark scores, prices, context windows and speed. The link updates as you pick, so you can share the exact comparison.
Cheapest blended
Claude Haiku 4.5
$2.00 / 1M tokens
Largest context
GPT-5
400K tokens
Fastest output
Claude Haiku 4.5
91 tokens/s
Top on SWE-bench Verified
Claude Sonnet 4.5
77.2%
Specifications
| Spec | Claude Haiku 4.5Anthropic | GPT-5OpenAI | Claude Sonnet 4.5Anthropic |
|---|---|---|---|
| Provider | Anthropic | OpenAI | Anthropic |
| Released | 15 Oct 2025 | 7 Aug 2025 | 29 Sep 2025 |
| Licence | Proprietary | ProprietaryRetires 11 Dec 2026 | ProprietaryRetires 30 Nov 2026 |
| Context window | 200K tokens | 400K tokens | 200K tokens |
| Max output | 64K tokens | 128K tokens | 64K tokens |
| Input / 1M | $1.00* | $1.25* | $3.00* |
| Output / 1M | $5.00 | $10.00 | $15.00 |
| Cached input / 1M | $0.10 | $0.125 | $0.30 |
| Speed | 91 tokens/s | 75 tokens/s | – |
| Input types | text, image | text, image | text, image |
| Best for | BudgetCoding | Coding | CodingAgents |
* Claude Haiku 4.5: Cheapest current Claude model. Uses the previous tokenizer. Minimum cacheable prompt: 4,096 tokens. Retirement commitment is only 'not sooner than 15 Oct 2026'; no deprecation notice yet.
* GPT-5: There is no separate cache-write charge. Batch and Flex: $0.625 / $5. Fast mode: $2.50 / $20. Snapshot gpt-5-2025-08-07 shuts down 11 Dec 2026; the recommended replacement is gpt-5.6-sol.
* Claude Sonnet 4.5: Deprecated 30 Sep 2026; retires on the Claude API on 30 Nov 2026. Recommended replacement: Claude Sonnet 5.5 ($2/$10). Uses the previous tokenizer. Minimum cacheable prompt: 1,024 tokens.
Benchmark scores
Officially reported scores only. A model with no published score on a benchmark shows a line, not a zero.
Price per 1M tokens
Context window and speed
Notes
- Claude Haiku 4.5
- Anthropic's fastest model with near-frontier intelligence; Anthropic positions it as giving coding performance similar to Claude Sonnet 4 at one-third the cost and more than twice the speed.
- GPT-5
- The previous-generation reasoning model for coding and agentic tasks, launched August 2025. OpenAI now points users to newer models.
- Claude Sonnet 4.5
- Deprecated Sonnet model; at launch Anthropic called it the best coding model in the world and the strongest model for building complex agents.
Still torn between two?
Test them on your own prompts.
Benchmarks are someone else’s work. On a free call we look at yours and agree how to check quality, cost and latency for the models you are weighing.