Models / GPT-4.1

OpenAI

GPT-4.1

28 published results from 4 sources. Each card shows where the number comes from and what it does not measure. Results are never combined into one score.

Provider
OpenAI
Sources
4
Our benchmarks
0
Price
Not yet published

Reported by others

Reported by Arena (formerly LMArena)
1,408
Overall · rank 127 of 177
Unit
Arena rating, higher is better
Range
1,402 to 1,415
Sample
7827 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.
Reported by Arena (formerly LMArena)
1,415
Business, management and finance · rank 107 of 402
Unit
Arena rating, higher is better
Range
1,408 to 1,422
Sample
8215 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,402
Creative writing · rank 79 of 407
Unit
Arena rating, higher is better
Range
1,395 to 1,410
Sample
6599 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,405
Expert prompts · rank 136 of 359
Unit
Arena rating, higher is better
Range
1,394 to 1,417
Sample
2592 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,403
Instruction following · rank 111 of 409
Unit
Arena rating, higher is better
Range
1,398 to 1,409
Sample
13204 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,415
Overall · rank 118 of 409
Unit
Arena rating, higher is better
Range
1,411 to 1,419
Sample
50001 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,405
Writing, literature and language · rank 91 of 408
Unit
Arena rating, higher is better
Range
1,399 to 1,412
Sample
11250 votes
Configuration
GPT-4.1
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
23.9%
Memory · rank 35 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not long-term personal memory in a product: sessions are BFCL's scripted ones.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
21.5%
Memory · rank 41 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not long-term personal memory in a product: sessions are BFCL's scripted ones.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
38.9%
Multi-turn tasks · rank 29 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not open-ended agent work: the tools and tasks are BFCL's simulated APIs.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
9.8%
Multi-turn tasks · rank 70 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not open-ended agent work: the tools and tasks are BFCL's simulated APIs.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
54.0%
Overall accuracy · rank 20 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not a neutral average: the weighting is BFCL's. Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
39.4%
Overall accuracy · rank 45 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not a neutral average: the weighting is BFCL's. Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
87.5%
Relevance detection · rank 27 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not whether the call itself was right; only that one was attempted.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
100.0%
Relevance detection · rank 1 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not whether the call itself was right; only that one was attempted.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
82.8%
Single-turn calls (curated) · rank 60 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
88.7%
Single-turn calls (curated) · rank 12 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
70.0%
Single-turn calls (user-contributed) · rank 59 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
78.9%
Single-turn calls (user-contributed) · rank 18 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
68.0%
Web search · rank 18 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not general research quality: questions have short, checkable answers.
Reported by Berkeley Function Calling Leaderboard (BFCL) V4
35.0%
Web search · rank 32 of 109
Unit
% correct, higher is better
Configuration
gpt-4.1
Measured
16 Dec 2025
Not shown
Not general research quality: questions have short, checkable answers.
Reported by UGI Leaderboard
6.0%
Requested-length error · rank 41 of 370
Unit
% off the requested word count, lower is better
Configuration
GPT-4.1
Measured
24 Sep 2025
Not shown
Not other format limits such as character counts or bullet counts.
Reported by UGI Leaderboard
0.33
Style adherence · rank 266 of 370
Unit
score from 0 to 1, higher is better
Configuration
GPT-4.1
Measured
24 Sep 2025
Not shown
Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
Reported by UGI Leaderboard
67.5
Writing score · rank 44 of 370
Unit
score out of 100, higher is better
Configuration
GPT-4.1
Measured
24 Sep 2025
Not shown
Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
Reported by Vectara Hallucination Leaderboard
99.9%
Answer rate · rank 19 of 108
Unit
% of documents, higher is better
Configuration
GPT-4.1
Measured
22 Sep 2026
Not shown
Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.
Reported by Vectara Hallucination Leaderboard
5.6%
Hallucination rate · rank 21 of 108
Unit
% of summaries, lower is better
Configuration
GPT-4.1
Measured
22 Sep 2026
Not shown
Not errors in open questions or other tasks: only summarisation, judged by Vectara's own model (HHEM-2.3), not by people, on news-style documents rather than your data.

Compare GPT-4.1 with