Models, measured side by side
Claude Sonnet 4 vs GPT-4.1
Compare Claude Sonnet 4, GPT-4.1 by token pricing, context length, benchmark results, speed and tool support.
Monthly workload
Monthly cost breakdown
InputOutput
Claude Sonnet 4
$135 / mo
GPT-4.1
$80 / mo
Performance comparison
Claude Sonnet 4
16.6
GPT-4.1
12.7
Overview
Pricing & capacity
| Pricing & capacity | Claude Sonnet 4 | GPT-4.1 |
|---|---|---|
| Input / 1M tokens | $3.00 | $2.00 |
| Output / 1M tokens | $15 | $8.00 |
| Cached input / 1M | $0.30 | $0.50 |
| Monthly cost | $135 | $80 |
| Context window | 200K | 1M |
| Maximum output | 64K | 33K |
Input / 1M tokens
- Claude Sonnet 4
- $3.00
- GPT-4.1
- $2.00
Output / 1M tokens
- Claude Sonnet 4
- $15
- GPT-4.1
- $8.00
Cached input / 1M
- Claude Sonnet 4
- $0.30
- GPT-4.1
- $0.50
Monthly cost
- Claude Sonnet 4
- $135
- GPT-4.1
- $80
Context window
- Claude Sonnet 4
- 200K
- GPT-4.1
- 1M
Maximum output
- Claude Sonnet 4
- 64K
- GPT-4.1
- 33K
Benchmarks & performance
| Benchmarks & performance | Claude Sonnet 4 | GPT-4.1 |
|---|---|---|
| Arena Elo | 1,250 | 1,200 |
| MMLU-Pro | 81.9% | 73.6% |
| Intelligence Index | 16.6 | 12.7 |
| GPQA | 68.3% | 66.6% |
| Humanity’s Last Exam | 4.3% | 4.2% |
| SciCode | Not listed | Not listed |
| Output speed | Not listed | 166 tokens/s |
| Time to first token | Not listed | 0.83 s |
| DesignArena rating | 1,145 | 1,029 |
| DesignArena win rate | 53.4% | 50.9% |
Arena Elo
- Claude Sonnet 4
- 1,250
- GPT-4.1
- 1,200
MMLU-Pro
- Claude Sonnet 4
- 81.9%
- GPT-4.1
- 73.6%
Intelligence Index
- Claude Sonnet 4
- 16.6
- GPT-4.1
- 12.7
GPQA
- Claude Sonnet 4
- 68.3%
- GPT-4.1
- 66.6%
Humanity’s Last Exam
- Claude Sonnet 4
- 4.3%
- GPT-4.1
- 4.2%
SciCode
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Not listed
Output speed
- Claude Sonnet 4
- Not listed
- GPT-4.1
- 166 tokens/s
Time to first token
- Claude Sonnet 4
- Not listed
- GPT-4.1
- 0.83 s
DesignArena rating
- Claude Sonnet 4
- 1,145
- GPT-4.1
- 1,029
DesignArena win rate
- Claude Sonnet 4
- 53.4%
- GPT-4.1
- 50.9%
6 of 10
Tools & features
| Tools & features | Claude Sonnet 4 | GPT-4.1 |
|---|---|---|
| Function calling | Supported | Supported |
| Structured outputs | Not listed | Supported |
| JSON mode | Not listed | Supported |
| Reasoning | Supported | Not listed |
| Built-in web search | Not listed | Not listed |
| Log probabilities | Not listed | Not listed |
| Deterministic seed | Not listed | Supported |
| Parallel tool calls | Not listed | Not listed |
| Prompt caching | Supported | Supported |
Function calling
- Claude Sonnet 4
- Supported
- GPT-4.1
- Supported
Structured outputs
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Supported
JSON mode
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Supported
Reasoning
- Claude Sonnet 4
- Supported
- GPT-4.1
- Not listed
Built-in web search
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Not listed
Log probabilities
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Not listed
Deterministic seed
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Supported
Parallel tool calls
- Claude Sonnet 4
- Not listed
- GPT-4.1
- Not listed
Prompt caching
- Claude Sonnet 4
- Supported
- GPT-4.1
- Supported
6 of 9
API & availability
| API & availability | Claude Sonnet 4 | GPT-4.1 |
|---|---|---|
| API identifier | claude-sonnet-4 | gpt-4.1 |
| API providers | Amazon Bedrock, Amazon Bedrock | Azure, OpenAI, Azure |
| Knowledge cutoff | 2025-01-31 | 2024-06-30 |
| Prices checked | Oct 8, 2026 | Oct 8, 2026 |
API identifier
- Claude Sonnet 4
claude-sonnet-4- GPT-4.1
gpt-4.1
API providers
- Claude Sonnet 4
- Amazon Bedrock, Amazon Bedrock
- GPT-4.1
- Azure, OpenAI, Azure
Knowledge cutoff
- Claude Sonnet 4
- 2025-01-31
- GPT-4.1
- 2024-06-30
Prices checked
- Claude Sonnet 4
- Oct 8, 2026
- GPT-4.1
- Oct 8, 2026
About the models
GPT-4.1
Non-reasoning GPT-4.1 with a 1M-token context window, strong at instruction following and coding.
Full pricing & detailsComparison FAQ
Claude Sonnet 4: $3.00 input and $15 output per million tokens. GPT-4.1: $2.00 input and $8.00 output per million tokens. The cheaper choice depends on your input-to-output ratio; a model can have a lower input rate but a higher output rate.
