Claude 3.5 Haiku is 73% cheaper than Claude 3.5 Sonnet (v2) on blended 3:1 token pricing. Claude 3.5 Haiku costs $0.80/1M in and $4.00/1M out, while Claude 3.5 Sonnet (v2) costs $3.00/1M in and $15.00/1M out.
Claude 3.5 Haiku vs Claude 3.5 Sonnet (v2)
Compare verified input/output token rates, prompt caching discounts, latency tiers, and context limits between Anthropic and Anthropic.
| Metric / Feature | Claude 3.5 Haiku (Anthropic) | Claude 3.5 Sonnet (v2) (Anthropic) |
|---|---|---|
| Input Price ($ / 1M Tokens) | $0.80 | $3.00 |
| Output Price ($ / 1M Tokens) | $4.00 | $15.00 |
| Prompt Caching Input Rate | $0.080/1M | $0.300/1M |
| Blended 3:1 Rate (Production) | $1.600 / 1M | $6.000 / 1M |
| Max Context Window | 128k | 200k |
| Multimodal Vision | ❌ Text Only | ✅ Supported |
| MMLU Benchmark | 75.2% | 88.3% |
| Throughput & Latency | Ultra-Fast (~95 t/s) | Fast (~50 t/s) |
| Best For | Real-time agents, code completion, fast document scanning | Autonomous coding, complex tool use, multi-step agent reasoning |
When to Choose Claude 3.5 Haiku
Choose Claude 3.5 Haiku if your workload requires real-time agents, code completion, fast document scanning. Ideal for teams needing Anthropic's infrastructure and ecosystem tooling.
When to Choose Claude 3.5 Sonnet (v2)
Choose Claude 3.5 Sonnet (v2) if your primary objective is autonomous coding, complex tool use, multi-step agent reasoning. Ideal for scaling high-throughput pipelines with Anthropic.
Want to compare Claude 3.5 Haiku against another model?
Select another target model to view immediate pricing & latency trade-offs.