GPT-5 vs Claude Sonnet 4.6
Side-by-side comparison of GPT-5 (OpenAI) and Claude Sonnet 4.6 (Anthropic). Exact API pricing per million tokens, context windows, output speed, and total cost on real-world prompts.
Specifications
| Spec | GPT-5 | Claude Sonnet 4.6 |
|---|---|---|
| Provider | OpenAI | Anthropic |
| Model id | gpt-5 | claude-sonnet-4.6 |
| Input price (per 1M tokens) | $1.25 | $3.00 |
| Output price (per 1M tokens) | $10.00 | $15.00 |
| Context window | 400,000 | 1,000,000 |
| Output speed (tokens/sec) | ~110 | ~90 |
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 | GPT-5 | Claude Sonnet 4.6 | Cheaper |
|---|---|---|---|---|---|
| Short question + answer | 50 | 150 | $0.001563 | $0.002400 | GPT-5 |
| Code review on one file | 500 | 1,500 | $0.015625 | $0.024000 | GPT-5 |
| Long document summary | 5,000 | 500 | $0.011250 | $0.022500 | GPT-5 |
| Heavy reasoning task | 2,000 | 8,000 | $0.082500 | $0.126000 | GPT-5 |
| Full codebase analysis | 50,000 | 10,000 | $0.162500 | $0.300000 | GPT-5 |
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 GPT-5 for
- •output-heavy workloads (long-form generation, code, summaries) — gpt-5 is meaningfully cheaper per output token
- •input-heavy workloads (long context, RAG, document QA) — gpt-5 is cheaper per input token
Pick Claude Sonnet 4.6 for
- •tasks needing a larger context window — claude-sonnet-4.6 fits 3x more tokens than gpt-5
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 GPT-5 to Claude Sonnet 4.6
# Before
model = "gpt-5"
# After
model = "claude-sonnet-4.6"If the providers differ (OpenAI vs Anthropic), 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 →