Gemini 3.8 Flash vs Claude Haiku 5.5: API Pricing Compared
Official prices, verified 2026-10-24. Every conclusion on this page is computed from the price data — if the prices change, the verdict changes.
| Gemini 3.8 Flash (Google) | Claude Haiku 5.5 (Anthropic) | |
|---|---|---|
| Input / 1M | $0.75 | $0.1 |
| Output / 1M | $3.75 | $0.5 |
| Cached input / 1M | Not published | $0.01 |
| Batch tier | Not published | $0.05 / $0.25 |
| Context window | Not published | 1,048,576 |
| Price verified | 2026-10-24 | 2026-10-24 |
Monthly cost at three workloads
| Workload | Gemini 3.8 Flash | Claude Haiku 5.5 |
|---|---|---|
| Chat assistant 2,000 input / 500 output tokens per call, 10,000 calls a month | $33.75 | $4.50 |
| Document processing 50,000 input / 2,000 output tokens per call, 1,000 calls a month | $45.00 | $6.00 |
| Agent loop with prompt caching 10,000 fresh input / 100,000 cached input / 2,000 output tokens per call, 500 calls a month | $7.50 | $1.50 |
The verdict, from the data
- Claude Haiku 5.5 has cheaper input: $0.1 vs $0.75 per 1M tokens (87% less).
- Claude Haiku 5.5 has cheaper output: $0.5 vs $3.75 per 1M tokens (87% less).
- At a blended workload of 1M input + 250K output tokens, Claude Haiku 5.5 wins: $0.225 vs $1.69 per call batch.
- Only Claude Haiku 5.5 publishes a batch tier ($0.05 / $0.25), which halves its rates for latency-tolerant work.
- Only Claude Haiku 5.5 publishes a cached-input rate ($0.01 per 1M), which matters for prompt-heavy agent workloads.
Run your own numbers in the homepage calculator, or open the dedicated calculators for Gemini 3.8 Flash and Claude Haiku 5.5.