High-volume, cost-sensitive work
Choose Claude Sonnet 5.5
Claude Sonnet 5.5 costs 29% less on a typical mix of three input tokens for every output token ($4.00 against $5.63 per million tokens).
Claude Sonnet 5.5 vs GPT-5.4: price, context and capabilities
Choose Claude Sonnet 5.5 for high-volume, cost-sensitive work, writing and code generation, reused prompts and documents (RAG, agents), the latest release and the most recent knowledge; on the other points the two are level or the data is missing.
In short
- Claude Sonnet 5.5 is 29% cheaper than GPT-5.4 at the headline price ($4.00 / 1M tokens vs $5.63 / 1M tokens).
- GPT-5.4 reads more per request: 1.05M tokens against 1M.
- Claude Sonnet 5.5 is the newer model (September 2026, against Mar 5, 2026).
Which should you choose?
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Writing and code generation
Choose Claude Sonnet 5.5
Claude Sonnet 5.5 charges $10.00 per million output tokens, against $15.00 for GPT-5.4. Output dominates the bill when answers are long.
Reused prompts and documents (RAG, agents)
Choose Claude Sonnet 5.5
Claude Sonnet 5.5 charges $0.10 per million cached input tokens, against $0.25 for GPT-5.4.
Reasoning over long documents
Either one
They are practically level on Long-context reasoning (AA-LCR) (82.7% and 82%).
Latest release
Choose Claude Sonnet 5.5
Claude Sonnet 5.5 is the newer release (September 2026, against Mar 5, 2026 for GPT-5.4).
Most recent knowledge
Choose Claude Sonnet 5.5
Claude Sonnet 5.5 was trained on data up to June 2026, against August 2025 for GPT-5.4.
Monthly cost at three volumes
| Volume | Claude Sonnet 5.5 | GPT-5.4 | Difference |
|---|---|---|---|
| Prototype: 1M input, 250K output tokens | $4.50 | $6.25 | Claude Sonnet 5.5, $1.75 less |
| Production: 20M input, 5M output tokens | $90.00 | $125.00 | Claude Sonnet 5.5, $35.00 less |
| Scale: 500M input, 100M output tokens | $2,000 | $2,750 | Claude Sonnet 5.5, $750.00 less |
Standard rates, without caching or batch discounts. Volumes are per month.
Specifications side by side
| Claude Sonnet 5.5 | GPT-5.4 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Category | Language models | Language models |
| Status | Generally available | Legacy |
| Headline price | $4.00 / 1M tokens | $5.63 / 1M tokens |
| Input | $2.00 / 1M tokens | $2.50 / 1M tokens |
| Output | $10.00 / 1M tokens | $15.00 / 1M tokens |
| Cached input | $0.10 / 1M tokens | $0.25 / 1M tokens |
| Context window | 1M | 1.05M |
| Max output | 128k | 128k |
| Released | September 2026 | Mar 5, 2026 |
| Knowledge cutoff | June 2026 | August 2025 |
| Licence | Proprietary | Proprietary |
| Capabilities | Vision, Reasoning, Tool calling, Structured output, Prompt caching, Batch API | Vision, Reasoning, Tool calling, Structured output, Prompt caching, Batch API |
| API model id | claude-sonnet-5-5 | gpt-5.4 |
Benchmarks side by side
Leader in bold.
| Benchmark | Claude Sonnet 5.5 | GPT-5.4 |
|---|---|---|
| Humanity's Last Exam | 55% | 43.7% |
| Long-context reasoning (AA-LCR) | 82.7% | 82% |
| Intelligence Index | 56 | No data |
| SciCode | 61% | No data |
| Terminal-Bench Hard | No data | 57.6% |
| GPQA Diamond | No data | 92% |
| IFBench (instruction following) | No data | 73.9% |
| τ²-Bench (agentic tool use) | No data | 87.1% |
| MMMU-Pro (visual reasoning) | No data | 78.4% |
Independent evaluations by Artificial Analysis, run the same way for every model, at the reasoning setting shown. Percentages unless marked as an index.
Frequently asked questions
Should I choose Claude Sonnet 5.5 or GPT-5.4?
Choose Claude Sonnet 5.5 for high-volume, cost-sensitive work, writing and code generation, reused prompts and documents (RAG, agents), the latest release and the most recent knowledge; on the other points the two are level or the data is missing.
Which is cheaper, Claude Sonnet 5.5 or GPT-5.4?
Claude Sonnet 5.5 is 29% cheaper than GPT-5.4 at the headline price ($4.00 / 1M tokens vs $5.63 / 1M tokens).
Which has the larger context window, Claude Sonnet 5.5 or GPT-5.4?
GPT-5.4 reads more per request: 1.05M tokens against 1M.
What is the difference between Claude Sonnet 5.5 and GPT-5.4?
Claude Sonnet 5.5 is made by Anthropic and GPT-5.4 by OpenAI. Claude Sonnet 5.5 is 29% cheaper than GPT-5.4 at the headline price ($4.00 / 1M tokens vs $5.63 / 1M tokens). GPT-5.4 reads more per request: 1.05M tokens against 1M. Claude Sonnet 5.5 is the newer model (September 2026, against Mar 5, 2026).
How much do Claude Sonnet 5.5 and GPT-5.4 cost per month?
At 20 million input and 5 million output tokens a month, Claude Sonnet 5.5 costs $90.00 and GPT-5.4 costs $125.00, at standard rates.
Which is newer, Claude Sonnet 5.5 or GPT-5.4?
Claude Sonnet 5.5 was released on September 2026 and GPT-5.4 on Mar 5, 2026.