OpenAI model product
GPT-4.1
1 is OpenAI's latest and most advanced flagship model, significantly improving upon GPT-4 Turbo in performance across benchmarks, speed, and cost-effectiveness.
Updated Aug 12, 2026. Default version: GPT-4.1
LLMBoard score31.9GPT-4.1
Coverage100%28 benchmark families
Context window1MTokens
Official input price$2OpenAI API
GPT-4.1 Specifications
Technical details for the model's default version.
- Version
- GPT-4.1
- Released
- Apr 14, 2025
- Knowledge cutoff
- Jun 1, 2024
- Parameters
- N/A
- Context window
- 1M
- Max output
- 32.8K
- Inputs
- image, pdf, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-4.1 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-4.1 Benchmark Results
Benchmark scores for GPT-4.1.
| Video-MME (long, no subtitles) | 72.0% | 01 | 1 | 100.0% | C | |
| ComplexFuncBench | 65.5% | 02 | 7 | 83.3% | C | |
| OpenAI-MRCR: 2 needle 1M | 46.3% | 03 | 5 | 50.0% | C | |
| Internal API instruction following (hard) | 49.1% | 04 | 7 | 50.0% | C | |
| OpenAI-MRCR: 2 needle 128k | 57.2% | 04 | 9 | 62.5% | C | |
| Graphwalks parents >128k | 25.0% | 05 | 7 | 33.3% | C | |
| Aider-Polyglot Edit | 52.9% | 06 | 10 | 44.4% | C | |
| COLLIE | 65.8% | 06 | 10 | 44.4% | C | |
| Graphwalks BFS <128k | 61.7% | 07 | 11 | 40.0% | C | |
| CharXiv-D | 87.9% | 08 | 16 | 53.3% | C | |
| Graphwalks parents <128k | 58.0% | 08 | 11 | 30.0% | C | |
| Graphwalks BFS >128k | 19.0% | 09 | 11 | 20.0% | C | |
| MMLU | 90.2% | 09 | 100 | 91.9% | C | |
| MathVista | 72.2% | 11 | 39 | 73.7% | C | |
| TAU-bench Airline | 49.4% | 13 | 23 | 45.5% | C | |
| TAU-bench Retail | 68.0% | 14 | 25 | 45.8% | C | |
| Aider-Polyglot | 51.6% | 15 | 22 | 33.3% | C | |
| Multi-IF | 70.8% | 15 | 20 | 26.3% | C | |
| MMMLU | 87.3% | 20 | 49 | 60.4% | C | |
| MMMU | 74.8% | 23 | 63 | 64.5% | C | |
| Multi-Challenge | 38.3% | 25 | 29 | 14.3% | C | |
| HMMT 2025 | 28.9% | 33 | 33 | 0.0% | C | |
| IFEval | 87.4% | 33 | 65 | 50.0% | C | |
| CharXiv-R | 56.7% | 39 | 48 | 19.1% | C | |
| AIME 2024 | 48.1% | 48 | 53 | 9.6% | C | |
| SWE-Bench Verified | 54.6% | 85 | 105 | 19.2% | C | |
| Humanity's Last Exam | 5.4% | 87 | 93 | 6.5% | C | |
| AIME 2025 | 46.4% | 106 | 114 | 7.1% | C | |
| GPQA | 66.3% | 144 | 234 | 38.6% | C |
GPT-4.1 Arena Results
Preference and agent-evaluation results for the default version.
| vision style control | overall | 64 | 1214.0 | 41,556 | N/A | |
| vision | overall | 68 | 1210.1 | 41,556 | N/A | |
| text style control | overall | 121 | 1414.1 | 51,136 | N/A | |
| text factuality | overall | 125 | 1408.5 | 7,857 | N/A | |
| text | overall | 146 | 1382.4 | 51,136 | N/A |
GPT-4.1 Runtime Performance
Provider-specific output speed and catalog latency for GPT-4.1. Runtime does not affect the capability score.
| OpenAI | 100 tok/s | 10 s | 1M | 32.8K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-4.1 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $2 input, $8 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- From $1.8 input, $7.2 output per 1M via Poe
- Tracked offerings
- 13
| Poe | openai/gpt-4.1 | global | $1.8 | $7.2 | 1M | |
| Impossibl | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| NanoGPT | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| FastRouter | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| NEAR AI Cloud | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| Ofox | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| OrcaRouter | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| Kilo Gateway | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| Merge Gateway | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| Vercel AI Gateway | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| OpenRouter | openai/gpt-4.1 | global | $2 | $8 | 1M | |
| OpenAI | gpt-4.1 | global | $2 | $8 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
GPT-4.1 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| GPT-4.1 | 31.9 | N/A | 1M | 32.8K | No | Proprietary |
What is GPT-4.1?
Key information about GPT-4.1 and its available data.
1 is OpenAI's latest and most advanced flagship model, significantly improving upon GPT-4 Turbo in performance across benchmarks, speed, and cost-effectiveness.
Data as of 2026-08-11.
GPT-4.1 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to GPT-4.1
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-5-nano
OpenAI
OP
o1 pro
OpenAI
OP
GPT-4.5
OpenAI
OP
o3 mini
OpenAI
OP
o1
OpenAI
OP
GPT-4.1-mini
OpenAI
FAQ
Common questions about GPT-4.1.
When was GPT-4.1 released?
GPT-4.1's default version was released on Apr 14, 2025.
How much does GPT-4.1 cost?
GPT-4.1's official API price is $2 per million input tokens and $8 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $1.8 input and $7.2 output via Poe.
Who created GPT-4.1?
GPT-4.1 was created by OpenAI.
What is the context window for GPT-4.1?
The default version has a 1M token context window.
Is GPT-4.1 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-4.1?
13 provider offerings are linked to the default version.
What models should I compare GPT-4.1 with?
Nearby ranked alternatives include Nemotron Nano 9B, Claude Sonnet 3.5, Qwen3 30B A3B.