GPT-4.1

GPT-4.1 is a non-reasoning model from OpenAI. 30 benchmarks count toward its score, in 8 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.38.5 ±3.0
CoverageShare of the index weight with results.100%
SpeedOutput tokens per second.53/s
Input / 1MUS dollars per 1M input tokens.$2 batch $1
Output / 1MUS dollars per 1M output tokens.$8 batch $4 US dollars per 1M output tokens in a batch.
ContextMaximum tokens in one request.1.05M
EloLMArena rating and rank.1383 (#157)

The index is a score out of 100. The ± range shows how much it can change. Batch work costs less.

49,941 votes. Elo shows what people prefer. It does not change the score.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
36.1
CodingCode writing and repair.
33.4
ReasoningLogic problems and puzzles.
39.0
MultimodalTasks with images and text.
43.5
KnowledgeFacts and expert knowledge.
41.4
MultilingualTasks in many languages.
71.9
InstructionTasks with strict rules in the prompt.
36.2
MathMath problems.
36.9

Results

30 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
Massive Multitask Language UnderstandingKnowledge90.2%Dan Hendrycks et al.
MGSMMultilingual87.7%9 Jan 2026Vals AI
Instruction-Following EvalInstruction87.4%39.6Jeffrey Zhou et al.
MATH 500Math87.2%39.39 Jan 2026Vals AI
MATH level 5Math83.0%39.5Epoch AI
MMLU ProKnowledge80.5%47.31 Sept 2026Vals AI
τ²-bench RetailAgentic74.0%51.9Sierra30 Apr 2026Sierra Research
MMMU ProMultimodal72.4%45.11 Sept 2026Vals AI
Artificial Analysis Long Context ReasoningReasoning68.3%55.5Artificial Analysis
GPQA diamondKnowledge66.9%40.0Epoch AI
Artificial Analysis GPQA DiamondKnowledge66.6%36.8Artificial Analysis
Graduate-Level Google-Proof Q&AKnowledge66.3%39.5David Rein et al.
GPQA DiamondKnowledge65.4%38.61 Sept 2026Vals AI
EuroEval SwedishMultilingual63.8%82.0EuroEval
EuroEval DutchMultilingual62.6%80.5EuroEval
EuroEval FrenchMultilingual62.3%80.1EuroEval
Artificial Analysis MMMU-ProMultimodal61.2%41.9Artificial Analysis
EuroEval ItalianMultilingual57.1%73.6EuroEval
τ²-bench AirlineAgentic56.0%39.0Sierra2 Mar 2026Sierra Research
LiveCodeBenchCoding54.7%34.01 Sept 2026Vals AI
Software Engineering Benchmark VerifiedCoding54.6%41.3Carlos E. Jimenez et al.
EuroEval PortugueseMultilingual54.4%70.3EuroEval
EuroEval PolishMultilingual54.1%69.9EuroEval
EuroEval SpanishMultilingual52.8%68.2EuroEval
EuroEval GermanMultilingual51.9%67.1EuroEval
SWE-Bench verifiedCoding48.5%36.5Epoch AI
τ²-Bench Tool-Agent-User EvaluationAgentic47.1%32.6Victor Barres et al.
Artificial Analysis IFBenchInstruction43.0%32.7Artificial Analysis
AIMEMath39.6%32.216 Apr 2026Vals AI
SWE-bench VerifiedCoding39.6%29.3mini-SWE-agent26 Feb 2026SWE-bench team
SWE-bench VerifiedCoding39.6%29.3mini-SWE-agent1 Sept 2026SWE-bench team
OTIS Mock AIME 2024-2025Math38.3%33.0Epoch AI
τ²-bench TelecomAgentic34.0%23.2Sierra2 Mar 2026Sierra Research
Terminal-Bench 1.0Agentic33.8%40.212 Jan 2026Vals AI
SWE-bench MultimodalMultimodal31.1%GUIRepair17 Nov 2025SWE-bench team
SWE-bench MultimodalMultimodal31.1%GUIRepair17 Nov 2025SWE-bench team
SimpleQA VerifiedKnowledge31.1%50.3Epoch AI
Artificial Analysis Omniscience AccuracyKnowledge27.8%48.3Artificial Analysis
Gert Labs Composite Game BenchmarkAgentic25.6%35.4Gert Labs
Terminal-Bench 2.0Agentic14.6%33.14 Jun 2026Vals AI
Artificial Analysis Intelligence IndexKnowledge12.7%38.3Artificial Analysis
Chess PuzzlesReasoning6.0%31.3Epoch AI
FrontierMath-Tiers-1-3-v2-PrivateMath6.0%34.8Epoch AI
FrontierMath-2025-02-28-PrivateMath5.5%35.8Epoch AI
ARC-AGI-1 (semi-private)Reasoning5.5%29.8ARC Prize Foundation
Artificial Analysis Humanity's Last ExamKnowledge4.2%29.1Artificial Analysis
ARC-AGI-2 (semi-private)Reasoning0.4%39.9ARC Prize Foundation
Critical Physics TasksReasoning0.0%38.4Artificial Analysis
FrontierMath-Tier-4-2025-07-01-PrivateMath0.0%43.4Epoch AI

30 benchmarks count, from 45 of 49 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authorsOpenRouter, collected directlyNo licence statedVals AI, collected directlyNo licence stated. Read from the public leaderboard and credited to Vals AIEpoch AI, collected directlyCC BY — free to use and redistribute with attributionSierra Research, collected directlyMIT — results are in the licensed repositoryEuroEval, collected directlyMIT — the leaderboard site and its CSV routes are in the licensed repositorySWE-bench team, collected directlyNo licence stated for the leaderboard. The harness repo is MITARC Prize Foundation, collected directlyNo licence stated. Their terms ask for written permission before commercial use

Same level, lower price

Qwen3.8-Flash-Next66.4 · Freedots3-note Preview65.5 · FreeApodex 1.1 Mini59.1 · Free

More from OpenAI

GPT-6 Astra83.5GPT-5.6 Sol76.7GPT-5.6 Terra73.4GPT-6 Sol76.9GPT-5.572.3