At a glance
| Provider | Anthropic |
|---|---|
| Released | 2024-11 |
| Tier | fast / cheap |
| License | Closed / proprietary |
| Modalities | text |
| Context window | 200k tokens |
| Max output | 8.192k tokens |
| API price · input | $0.80 / 1M tokens |
| API price · output | $4.00 / 1M tokens |
Benchmark performance
How Claude 3.5 Haiku stacks up against the current leader (GPT-5) and the median model in the leaderboard:
| Benchmark | Claude 3.5 Haiku | GPT-5 | Median |
|---|---|---|---|
| Chatbot Arena Elo | 1240 | 1410 | 1320 |
| MMLU-Pro | 65.0 | 86.8 | 78.0 |
| GPQA Diamond | 41.6 | 87.3 | 65.0 |
| MATH | 69.4 | 96.7 | 78.3 |
| HumanEval | 88.1 | 95.1 | 92.0 |
| SWE-Bench Verified | 40.6 | 74.9 | 49.0 |
Numbers compiled from provider technical reports and Chatbot Arena. See methodology for the composite-score formula.
OpenRouter routes your requests across Claude 3.5 Haiku, GPT-5, Claude, Gemini, and 100+ other models behind a single API key — pay-as-you-go, no monthly minimum. Try OpenRouter → (affiliate · supports this site)
What does Claude 3.5 Haiku cost in practice?
API pricing is $0.80 per 1M input tokens and $4.00 per 1M output tokens. Assuming a 50/50 input/output split, here is what that looks like at three workload sizes:
| Volume | Per day | Per month | Per year |
|---|---|---|---|
| 1M tokens/day | $2.40 | $72.00 | $876.00 |
| 10M tokens/day | $24.00 | $720.00 | $8,760 |
| 100M tokens/day | $240.00 | $7,200 | $87,600 |
Strengths & weaknesses
Where it shines
No clear top-tier strengths — this is a mid-pack model.
Where it lags
- Arena: 1240 (rank #26 of 27, below average)
- MMLU-Pro: 65.0 (rank #27 of 29, below average)
- GPQA: 41.6 (rank #27 of 29, below average)
Best alternatives
The closest models to Claude 3.5 Haiku by tier and benchmark score:
| Model | Score | $ in / out | Context | Action |
|---|---|---|---|---|
| GPT-4o mini OpenAI |
61.3 | $0.15 / $0.60 | 128k | Try → · vs Claude |
| Gemini 2.0 Flash |
65.6 | $0.10 / $0.40 | 1M | Try → · vs Claude |
| o3-mini OpenAI |
72.7 | $1.10 / $4.40 | 200k | Try → · vs Claude |
| Gemini 2.5 Flash |
73.3 | $0.30 / $2.50 | 1M | Try → · vs Claude |
| GPT-5 mini OpenAI |
77.0 | $0.25 / $2.00 | 400k | Try → · vs Claude |
Frequently asked questions
Is Claude 3.5 Haiku a good model?
Claude 3.5 Haiku scores 56.2 on the llmrank.top composite (rank #28 of 30). It trails the frontier — for top performance, look at GPT-5 (86.0). Whether it's the right fit depends on your workload — see the use-case discussion on this page.
How much does Claude 3.5 Haiku cost?
Claude 3.5 Haiku is priced at $0.80 per 1M input tokens and $4.00 per 1M output tokens (USD). For a 10M-token-per-day workload split 50/50, that works out to roughly $24.00/day or $8,760/year.
What is Claude 3.5 Haiku's context window?
200k tokens. That covers most multi-document workloads.
Is Claude 3.5 Haiku open source?
No — it is a closed (proprietary) model accessed only via the provider API.
What is Claude 3.5 Haiku's SWE-Bench score?
Claude 3.5 Haiku scores 40.6% on SWE-Bench Verified — the benchmark that measures real-world GitHub issue resolution. Its Chatbot Arena Elo is 1240.
What are the best alternatives to Claude 3.5 Haiku?
Closest alternatives by tier and score: GPT-4o mini, Gemini 2.0 Flash, o3-mini. See the alternatives section on this page for side-by-side numbers.
Related: GPT-4o mini · Gemini 2.0 Flash · o3-mini · Full leaderboard
Spotted out-of-date numbers? Open an issue — corrections usually ship within 24h.
Try Claude 3.5 Haiku now
Direct link to the official playground — or use OpenRouter to A/B test it against any other model on a single API key.