GPT-4.1
openai/gpt-4.1- OpenAI
- Family · GPT-4
- Released 14 Apr 2025
- Image · Text · File → Text
- Tools · JSON · Vision
Not on OSMEXO yet
Spec & list price
Context1M
Max output32K
Input / 1M$2.00
Output / 1M$8.004× input
Scores · standing
Intelligence12.7#202/489
Usage
Tokens / day3.7B▼ 31% 7d
Usage rank#133
Requests / day740.0K
Tool calls / req0.08
Per request · 30 days
Avg input3.8K0.3× median
Avg output1470.2× median
Cache hit48%0.7× median
Reasoning share0%0.0× median
Tool-call failures0.6%0.3× median
Consumption
Peak 15.0B 15 OctLow 2.5B 1 NovRank today #133
News & trends
9 mentions this week ▲ 13%this model · videos, posts, articles · 30 days
Trending · 14 days
- OpenAI, API, GPT-4o, GPT-4.1, GPT-6 Luna, モデル提供終了, モデル移行, ファインチューニング, 開発者向け, AIニュース271 views
- 【実測】新gpt-4.1-miniに議事録要約を任せたら7秒→1秒331 views
- OpenAI is shutting down GPT-4o and GPT-4.1 on Jan 14. We tested the cheaper Luna models task by task first. Results were mixed.4 upvotes
News · latest from OpenAI
- GPT-6.1 Terra vs. Claude Opus 5.5: Cheaper and faster in three tests at maxthenewstack.io · 2 Nov
- Chinese labs now target OpenAI's GPT-6 Astra for distillationfork.zone · 1 Nov
- GPT-6.1 Terrasimonwillison.net · 31 Oct
- GPT-5.6 Sol can beat Factorio on peaceful moderyanmadden.net · 31 Oct
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 3 providers, latest day
Output speed#120/33482 tok/s1.2× faster than median
Fastest provider107 tok/sbest of 3 providers
First token#1831.18 s1.2× longer than median
#26 cheapest of 29 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases12.7
Technical details
- Tokenizer
- GPT
- Knowledge cutoff
- 2024-06-30
- Moderation
- moderated
Request parameters 9
- tools
- tool_choice
- response_format
- structured_outputs
- temperature
- top_p
- max_tokens
- max_completion_tokens
- seed
Interesting facts
- Less output room than similar models (32.8K against a 114.4K median).
- Released 567 days ago, the newest of 3 GPT-4 models.
- Weekdays average 9.8B tokens a day against 5.4B on weekends.
When to use
Best for
- Long document and codebase question answering
- Image and file understanding
- General knowledge and math assistants
- Reliable tool calling without reasoning delays
- Offline bulk jobs via the batch variant
Think twice for
- Hardest reasoning tasks
- Large single-response outputs
- Self-hosting (closed weights)
Research papers
- Concordat: Typed Sessions for LLM Agents27 Oct
- Retrospective Comparison of Radiation Oncology Patient Message Responses Generated by Epic and Institutionally Managed LLMs25 Oct
- Green Code or Costly Code? Evaluating the Energy Costs of Iterative LLM-Based Code Optimizations23 Oct
- Behind Aggregate Scores: What Systematically Varying Arena-Hard's LLM Judge Reveals21 Oct
- PhishGen: Understanding LLM Capabilities in Automated Phishing Website Generation21 Oct
- CARLAR Influence: A Structured Hallucination Risk in Multimodal Cheque Extraction and an Exploratory Compressed-Image Countermeasure19 Oct
- Towards More Granular Reasoning: Mapping the DSA Transparency Database Taxonomy to Platform Terms and Conditions Using Large Language Models19 Oct
- One Skill Takes Over: Clashing Agent Skills and a One-File Fix18 Oct
Same family
GPT-4- 6.7
- 9.6
Alternatives
other vendors · similar capability- 16.1
- 8.4
- 14.2
- 7.7
Want GPT-4.1 on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.






