Claude 3.5 Haiku vs Gemini 2.5 Flash
Side-by-side comparison of Claude 3.5 Haiku (Anthropic) and Gemini 2.5 Flash (Google). Exact API pricing per million tokens, context windows, output speed, and total cost on real-world prompts.
Specifications
| Spec | Claude 3.5 Haiku | Gemini 2.5 Flash |
|---|---|---|
| Provider | Anthropic | |
| Model id | claude-3.5-haiku | gemini-2.5-flash |
| Input price (per 1M tokens) | $0.80 | $0.30 |
| Output price (per 1M tokens) | $4.00 | $2.50 |
| Context window | 200,000 | 1,048,576 |
| Output speed (tokens/sec) | ~120 | ~200 |
Cost on real prompts
Total cost = (input tokens × input price) + (output tokens × output price). Numbers below use the exact pricing tables published by each provider.
| Scenario | Input | Output | Claude 3.5 Haiku | Gemini 2.5 Flash | Cheaper |
|---|---|---|---|---|---|
| Short question + answer | 50 | 150 | $0.00064 | $0.00039 | Gemini 2.5 Flash |
| Code review on one file | 500 | 1,500 | $0.006400 | $0.003900 | Gemini 2.5 Flash |
| Long document summary | 5,000 | 500 | $0.006000 | $0.002750 | Gemini 2.5 Flash |
| Heavy reasoning task | 2,000 | 8,000 | $0.033600 | $0.020600 | Gemini 2.5 Flash |
| Full codebase analysis | 50,000 | 10,000 | $0.080000 | $0.040000 | Gemini 2.5 Flash |
Want the exact cost for your prompt instead of these examples? Open the cost calculator pre-loaded with both models →
When to pick which
Heuristics derived from the spec table above. Always validate on your own prompts before committing — these are starting points, not verdicts.
Pick Claude 3.5 Haiku for
No clear advantage on the data points we measure. Compare on your actual prompts.
Pick Gemini 2.5 Flash for
- •output-heavy workloads — gemini-2.5-flash is meaningfully cheaper per output token
- •input-heavy workloads — gemini-2.5-flash is cheaper per input token
- •tasks needing a larger context window — gemini-2.5-flash fits 5x more tokens than claude-3.5-haiku
- •latency-sensitive UX — gemini-2.5-flash streams faster (~200 vs ~120 tok/s)
Switching between them
For most use cases, switching providers means updating the model id and the request shape if the providers differ. Within the same provider, it's usually a single-line change.
From Claude 3.5 Haiku to Gemini 2.5 Flash
# Before
model = "claude-3.5-haiku"
# After
model = "gemini-2.5-flash"If the providers differ (Anthropic vs Google), you'll also need to swap the SDK / endpoint URL. Cross-provider migrations usually take 30 minutes to a few hours depending on how many features (streaming, function calling, tool use) you depend on.
Calculate cost on your own prompt
The examples above use generic input/output ratios. For an exact comparison, paste your real prompt into the calculator — it counts tokens with the right tokenizer for each model and shows side-by-side cost.
Open the calculator with both models →