GPT-4.1 nano: benchmark scores, price and ranking
GPT-4.1 nano from OpenAI ranks #740 of 743 LLMs on the LLMs Tiger aggregate benchmark score (38.7), based on 12 benchmark results captured up to 2026-10-13.
- Organisation
- OpenAI
- Type
- Fast / low-cost (flash) model
- Open weights
- No
- Released
- 2025-04-14 (source: curated)
- Context window
- 1,047,576 tokens
- Cheapest API price
- $0.100 in / $0.400 out per million tokens (via azure, observed 2026-10-13T19:47Z)
- Aggregate rank
- #740 of 743 (score 38.7)
Relative to other models, GPT-4.1 nano is strongest on Epoch ECI (#91 of 204) and weakest on SimpleQA Verified (#69 of 69).
Across the 12 benchmarks where at least ten models are scored, GPT-4.1 nano is in the top 10% on 0 and in the top half on 1.
Its cheapest listed input price, $0.100 per million tokens, is lower than that of 79% of the 427 models with a listed price.
GPT-4.1 nano benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Epoch ECI | 130 | #91 of 204 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 28.9% | #124 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 48.9% | #191 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| CritPt | 0.0% | #94 of 115 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 52.5% | #123 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WeirdML | 19.0% | #96 of 109 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SciCode | 25.9% | #97 of 106 | Epoch AI Benchmarking Hub | 2026-10-13 |
| ARC-AGI-2 | 0.0% | #80 of 87 | ARC Prize official leaderboard | 2026-10-13 |
| Aider Polyglot | 8.9% | #47 of 49 | Aider polyglot leaderboard | 2026-10-13 |
| LMCA | 5.5% | #104 of 108 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Fiction.liveBench | 25.0% | #46 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleQA Verified | 6.0% | #69 of 69 | Epoch AI Benchmarking Hub | 2026-10-13 |
About GPT-4.1 nano
GPT-4.1 nano is a smaller, lower-cost model from OpenAI. It was released on 2025-04-14, according to the developer's announcement. It accepts up to 1,047,576 tokens of input at once. The cheapest API listing we track charges $0.100 per million input tokens and $0.400 per million output tokens, through azure. We hold 12 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where GPT-4.1 nano stands
On our aggregate score, GPT-4.1 nano is #740 of 743 ranked models, in the lower half. Its score is 38.7; the scale is centred on 50, so it sits 11.3 points below the typical model. That beats 0% of the ranked field. Among the 88 ranked OpenAI models it is #86. The next one up is GPT-4 (Jun 2023) at 41.2. The next one down is GPT-3.5 Turbo (Jan 2024) at 38.5.
GPT-4.1 nano, benchmark by benchmark
Epoch ECI puts GPT-4.1 nano at 130, #91 of 204. That beats 55% of the models tested, but 27 points short of Qwen 3.8 Max (156). The results either side are Llama 4 Scout (130) and Gemini 1.5 Flash (Sep 2024) (129). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so GPT-4.1 nano is 2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPT-4.1 nano lands below the middle on OTIS Mock AIME: 28.9%, #124 of 189, with 65% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is claude-sonnet-4-20250514 at 28.9%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so GPT-4.1 nano is 28.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
GPT-4.1 nano lands below the middle on GPQA Diamond: 48.9%, #191 of 275, with 69% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is mistral-large-2407 at 49.0%. It covers expert-level science questions. The board median is 71.7%, so GPT-4.1 nano is 22.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
CritPt is a weak spot. GPT-4.1 nano scores 0.0%, #94 of 115, behind 81% of the models measured. The leader, GPT-5.6 Sol (max), reaches 32.3%. It covers unpublished research-level physics problems. The board median is 2.9%, so GPT-4.1 nano is 2.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
DTBench is a weak spot. GPT-4.1 nano scores 52.5%, #123 of 146, behind 84% of the models measured. The leader, Claude Opus 5.5 (max), reaches 98.9%. It covers dynamic tool-use benchmark. The board median is 68.8%, so GPT-4.1 nano is 16.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
WeirdML is a weak spot. GPT-4.1 nano scores 19.0%, #96 of 109, behind 87% of the models measured. The leader, GPT-6 Astra (max), reaches 93.3%. It covers unusual machine-learning tasks. The board median is 41.2%, so GPT-4.1 nano is 22.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Aider Polyglot is a weak spot. GPT-4.1 nano scores 8.9%, #47 of 49, behind 94% of the models measured. The leader, gemini-2.5-pro-preview-06-05, reaches 83.1%. It covers edits across 6 languages (225 exercises). The board median is 44.9%, so GPT-4.1 nano is 36.0 points below it. Source: Aider polyglot leaderboard, captured 2026-10-13.
LMCA is a weak spot. GPT-4.1 nano scores 5.5%, #104 of 108, behind 95% of the models measured. The leader, Claude Opus 5.5 (max), reaches 68.2%. It covers language-model capability assessment. The board median is 29.2%, so GPT-4.1 nano is 23.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Fiction.liveBench is a weak spot. GPT-4.1 nano scores 25.0%, #46 of 47, behind 96% of the models measured. The leader, grok-4-0709, reaches 94.4%. It covers long-context story comprehension (16k). The board median is 61.1%, so GPT-4.1 nano is 36.1 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
SimpleQA Verified is a weak spot. GPT-4.1 nano scores 6.0%, #69 of 69, behind 99% of the models measured. The leader, gemini-4-argon-high, reaches 76.0%. It covers short factual questions. The board median is 41.0%, so GPT-4.1 nano is 35.0 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
The full table above also lists the other 2 results, including boards with fewer than ten models, where a rank says little.
Strengths and weaknesses
Software engineering: weak, ahead of 6% of the field on average across 2 benchmarks. Its best board here is SciCode (#97 of 106) and its weakest is Aider Polyglot (#47 of 49). Reasoning, maths and science: weak, ahead of 23% of the field on average across 4 benchmarks. Its best board here is OTIS Mock AIME (#124 of 189) and its weakest is ARC-AGI-2 (#80 of 87). General knowledge and chat: weak, ahead of 15% of the field on average across 5 benchmarks. Its best board here is Epoch ECI (#91 of 204) and its weakest is SimpleQA Verified (#69 of 69).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so GPT-4.1 nano costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on gpt-4o-mini-2024-07-18 ($0.150) and claude-3-haiku-20240307 ($0.250).
Release history
The OpenAI model released before it was GPT-4.1 mini on 2025-04-14, which scores 43.0. The next one was o3 (high) on 2025-04-16, at 49.8.
Reading these numbers
The 12 scores on this page come from 3 sources: Epoch AI Benchmarking Hub (10), ARC Prize official leaderboard (1), Aider polyglot leaderboard (1). 2 of them come from an official leaderboard and 10 from datasets or third-party runs.
Models ranked nearby
- #737 gemma-3-27b-it (39.7)
- #738 Meta-Llama-3-8B-Instruct (39.4)
- #739 gemma-3-12b-it (39.1)
- #741 GPT-3.5 Turbo (Jan 2024) (38.5)
- #742 gpt-4o-mini-2024-07-18 (38.3)
- #743 Llama-3.1-8B-Instruct (36.7)
More from OpenAI
- GPT-6 Astra (max) (#2, 64.0)
- GPT-6.1 Sol (max) (#5, 60.6)
- GPT-5.6 Sol (max) (#6, 60.1)
- GPT-6 Sol (max) (#7, 60.0)
- GPT-5.5 (xhigh) (#11, 58.0)
- GPT-5.6 Terra (max) (#12, 57.8)
Compare it on the live leaderboard → · How the aggregate score works