LLMs TigerLive LLM benchmarks

GPT-4.1: benchmark scores, price and ranking

GPT-4.1 from OpenAI ranks #620 of 743 LLMs on the LLMs Tiger aggregate benchmark score (44.9), based on 18 benchmark results captured up to 2026-10-13.

Organisation
OpenAI
Type
Frontier model
Open weights
No
Released
2025-04-14 (source: curated)
Context window
1,047,576 tokens
Cheapest API price
$2.00 in / $8.00 out per million tokens (via azure, observed 2026-10-13T19:47Z)
Aggregate rank
#620 of 743 (score 44.9)

Relative to other models, GPT-4.1 is strongest on Epoch ECI (#64 of 204) and weakest on FrontierMath T1-3 (#63 of 68).

Across the 17 benchmarks where at least ten models are scored, GPT-4.1 is in the top 10% on 0 and in the top half on 3.

Its cheapest listed input price, $2.00 per million tokens, is lower than that of 18% of the 427 models with a listed price.

GPT-4.1 benchmark results

GPT-4.1 benchmark scores, ranked against all models measured on each benchmark
BenchmarkScoreRankSourceCaptured
Epoch ECI137#64 of 204Epoch AI Benchmarking Hub2026-10-13
SWE-bench Multimodal31.1%#2 of 5SWE-bench Multimodal leaderboard2026-10-13
Aider Polyglot52.4%#20 of 49Aider polyglot leaderboard2026-10-13
Fiction.liveBench63.9%#20 of 47Epoch AI Benchmarking Hub2026-10-13
DTBench68.3%#75 of 146Epoch AI Benchmarking Hub2026-10-13
GPQA Diamond66.9%#150 of 275Epoch AI Benchmarking Hub2026-10-13
LMCA25.6%#63 of 108Epoch AI Benchmarking Hub2026-10-13
OTIS Mock AIME38.3%#114 of 189Epoch AI Benchmarking Hub2026-10-13
WeirdML39.0%#66 of 109Epoch AI Benchmarking Hub2026-10-13
Chess Puzzles6.0%#76 of 122Epoch AI Benchmarking Hub2026-10-13
EnigmaEval2.2%#15 of 24Epoch AI Benchmarking Hub2026-10-13
SimpleQA Verified31.1%#49 of 69Epoch AI Benchmarking Hub2026-10-13
ALE-Bench558#57 of 79Epoch AI Benchmarking Hub2026-10-13
SWE-bench Verified39.6%#55 of 73SWE-bench Verified leaderboard2026-10-13
Humanity's Last Exam5.4%#21 of 27Epoch AI Benchmarking Hub2026-10-13
SimpleBench27.0%#48 of 61Epoch AI Benchmarking Hub2026-10-13
ARC-AGI-20.4%#76 of 87ARC Prize official leaderboard2026-10-13
FrontierMath T1-36.0%#63 of 68Epoch AI Benchmarking Hub2026-10-13

About GPT-4.1

GPT-4.1 is a frontier 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 $2.00 per million input tokens and $8.00 per million output tokens, through azure. We hold 18 benchmark results for it across 4 areas, and none is in the top tenth of its board.

Where GPT-4.1 stands

On our aggregate score, GPT-4.1 is #620 of 743 ranked models, in the lower half. Its score is 44.9; the scale is centred on 50, so it sits 5.1 points below the typical model. That beats 17% of the ranked field. Among the 88 ranked OpenAI models it is #73. The next one up is GPT-3.5 Turbo (Nov 2023) at 45.4. The next one down is o1-mini-2024-09-12 at 44.8.

GPT-4.1, benchmark by benchmark

Epoch ECI puts GPT-4.1 at 137, #64 of 204. That beats 69% of the models tested, but 20 points short of Qwen 3.8 Max (156). The results either side are Qwen3-30B-A3B-Instruct (Jul 2025) (137) and GPT-4.5 (137). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so GPT-4.1 is 9 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

On Aider Polyglot, GPT-4.1 reaches 52.4%. It ranks #20 of 49, ahead of 59% of the field, and 30.7 points behind gemini-2.5-pro-preview-06-05 at 83.1%. The results either side are Optimus Alpha (52.9%) and Claude 3.5-Sonnet-20241022 (51.6%). It covers edits across 6 languages (225 exercises). The board median is 44.9%, so GPT-4.1 is 7.5 points above it. Source: Aider polyglot leaderboard, captured 2026-10-13.

GPT-4.1 manages 63.9% on Fiction.liveBench, good for #20 of 47. grok-4-0709 leads at 94.4%, 30.5 points further up, and 57% of models sit below. The results either side are Qwen3-Max-Instruct (66.7%) and GPT-4.5 Preview (Feb 2025) (63.9%). It covers long-context story comprehension (16k). The board median is 61.1%, so GPT-4.1 is 2.8 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

GPT-4.1 lands below the middle on DTBench: 68.3%, #75 of 146, with 51% of the field scoring higher. Claude Opus 5.5 (max) leads at 98.9%; the closest result above is o3-mini (high) at 68.8%. It covers dynamic tool-use benchmark. The board median is 68.8%, so GPT-4.1 is 0.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

GPT-4.1 lands below the middle on GPQA Diamond: 66.9%, #150 of 275, with 54% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is Llama 4 Maverick (FP8) at 67.0%. It covers expert-level science questions. The board median is 71.7%, so GPT-4.1 is 4.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

GPT-4.1 lands below the middle on LMCA: 25.6%, #63 of 108, with 57% of the field scoring higher. Claude Opus 5.5 (max) leads at 68.2%; the closest result above is Mistral Medium 3.5 at 26.1%. It covers language-model capability assessment. The board median is 29.2%, so GPT-4.1 is 3.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

GPT-4.1 lands below the middle on Humanity's Last Exam: 5.4%, #21 of 27, with 74% of the field scoring higher. Muse Spark leads at 40.6%; the closest result above is GPT-4.5 Preview (Feb 2025) at 5.4%. It covers expert-level questions across domains. The board median is 10.7%, so GPT-4.1 is 5.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

GPT-4.1 lands below the middle on SimpleBench: 27.0%, #48 of 61, with 77% of the field scoring higher. Claude Fable 5 (max) leads at 81.9%; the closest result above is gemini-1.5-pro-002 at 27.1%. It covers trick questions that test common sense. The board median is 41.4%, so GPT-4.1 is 14.4 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

ARC-AGI-2 is a weak spot. GPT-4.1 scores 0.4%, #76 of 87, behind 86% of the models measured. The leader, GPT-6 Astra (max), reaches 95.0%. It covers abstract reasoning, ARC Prize public eval set. The board median is 22.8%, so GPT-4.1 is 22.4 points below it. Source: ARC Prize official leaderboard, captured 2026-10-13.

FrontierMath T1-3 is a weak spot. GPT-4.1 scores 6.0%, #63 of 68, behind 91% of the models measured. The leader, GPT-6 Astra (max), reaches 93.7%. It covers research-level maths, tiers 1-3. The board median is 55.6%, so GPT-4.1 is 49.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

The full table above also lists the other 8 results, including boards with fewer than ten models, where a rank says little.

Strengths and weaknesses

Software engineering: uneven, ahead of 37% of the field on average across 3 benchmarks. Its best board here is Aider Polyglot (#20 of 49) and its weakest is SWE-bench Verified (#55 of 73). Reasoning, maths and science: weak, ahead of 27% of the field on average across 7 benchmarks. Its best board here is GPQA Diamond (#150 of 275) and its weakest is FrontierMath T1-3 (#63 of 68). General knowledge and chat: uneven, ahead of 46% of the field on average across 6 benchmarks. Its best board here is Epoch ECI (#64 of 204) and its weakest is SimpleQA Verified (#49 of 69).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so GPT-4.1 costs more 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 Command R+ ($2.50) and GPT-4o (May 2024) ($2.50).

Release history

The OpenAI model released before it was chatgpt-4o-latest (2025-03-29) on 2025-03-29, which scores 50.2. The next one was GPT-4.1 mini on 2025-04-14, at 43.0.

Reading these numbers

The 18 scores on this page come from 5 sources: Epoch AI Benchmarking Hub (14), ARC Prize official leaderboard (1), Aider polyglot leaderboard (1), SWE-bench Multimodal leaderboard (1). 4 of them come from an official leaderboard and 14 from datasets or third-party runs.

Models ranked nearby

More from OpenAI

Compare it on the live leaderboard → · How the aggregate score works