Name: GPT-4.1 Nano (2025-04-14)
Author: OpenAI

GPT-4.1 Nano (2025-04-14)

OpenAI · 2025-04-14

Immutable GPT-4.1 Nano API snapshot dated 2025-04-14 and evaluated on the Berkeley Function Calling Leaderboard v4.

Benchmark scores

Benchmark	Score
Arena Text — Business, Management & Financial Ops (No Style Control)	1266
Arena Text — Chinese (No Style Control)	1271
Arena Text — Creative Writing (No Style Control)	1261
Arena Text — English (No Style Control)	1303
Arena Text — Entertainment, Sports & Media (No Style Control)	1243
Arena Text — Exclude Ties (No Style Control)	1211
Arena Text — Exclude Ties (Style Controlled)	1267
Arena Text — Expert (No Style Control)	1271
Arena Text — German (No Style Control)	1288
Arena Text — Hard Prompts (No Style Control)	1286
Arena Text — Hard Prompts English (No Style Control)	1293
Arena Text — Instruction Following (No Style Control)	1267
Arena Text — Japanese (No Style Control)	1199
Arena Text — Legal & Government (No Style Control)	1300
Arena Text — Life, Physical & Social Science (No Style Control)	1297
Arena Text — Longer Query (No Style Control)	1283
Arena Text — Medicine & Healthcare (No Style Control)	1276
Arena Text — Multi-turn (No Style Control)	1277
Arena Text — Non-English (No Style Control)	1261
Arena Text — Non-English (Style Controlled)	1298
Arena Text — Occupational Mathematical (No Style Control)	1274
Arena Text — Russian (No Style Control)	1261
Arena Text — Software & IT Services (No Style Control)	1306
Arena Text — Writing, Literature & Language (No Style Control)	1261
Arena Vision — English (No Style Control)	1101
Arena Vision — English (Style Controlled)	1125
BFCL v4 Overall Accuracy	33.0
BigCodeBench-Hard Complete	31.8
BigCodeBench-Hard Instruct	28.4
LMArena Text Arena Chinese Elo	1321