BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases9
OpenAICompare with another model

GPT-4.5

GPT-4.5 is an OpenAI proprietary model, released 27 Feb 2025. Its best configuration ranks #388 of 785 on the BenchLeader Index at 49.7 ±4.9, in the upper half. The ± is the point: 323 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in human preference (63) and lowest in reasoning (40). Last measured 22 Oct 2026.

Blended price
–
Output speed
–
First answer
–
Context
128k
Full answer
–
Cost per run
–
Released
27 Feb 2025
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index50
  2. Reasoning40
  3. Coding50
  4. Agents & tools43
  5. Maths54
  6. Human preference63
  7. Multimodal53
  8. Long context55
  9. Composite40

Versions

OpenAI has shipped 13 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

ModelReleasedIndexRank
GPT-5.69 Jul 2026––
GPT-5.5xhigh23 Apr 202666.1#35
GPT-5.4xhigh5 Mar 202664.7#45
GPT-5.2xhigh11 Dec 202560.7#104
GPT-5.1high12 Nov 202557.7#170
GPT-5high7 Aug 202557.9#166
GPT-4.114 Apr 202548.3#428
GPT-4.5this page27 Feb 202549.7#388
GPT 4 012525 Jan 2024––
GPT-4 11066 Nov 2023––
GPT 4 061313 Jun 202339.7#677
GPT-414 Mar 202342.4#596
GPT-3.5–––

How its index has moved

8 Sept 2026 to 21 Oct 2026
455055

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmarknot statedSourceTrend
GPQA Diamond68.7%#173Epoch AI Benchmarking Hub
Humanity's Last Exam5.4%#41Scale AI / CAIS
SimpleBench34.5%#65SimpleBench
Arena Hard Prompts1440#125Arena (LMArena)
ARC-AGI-110.3%#235ARC Prize
ARC-AGI-20.8%#231ARC Prize
ForecastBenchnot in index64.8%#19ForecastBench

Coding

Benchmarknot statedSourceTrend
WeirdML39.4%#107WeirdML
Arena Coding1458#135Arena (LMArena)
Aider Polyglot44.9%#29Aider polyglot leaderboard

Agents & tools

Benchmarknot statedSourceTrend
Cybench17.5%#12Cybench

Maths

Benchmarknot statedSourceTrend
OTIS Mock AIME37.8%#200Epoch AI Benchmarking Hub
MATH Level 578.6%#35Epoch AI Benchmarking Hub
Arena Maths1408#139Arena (LMArena)

Instruction following

Benchmarknot statedSourceTrend
Arena Instruction Followingnot in index1437#84Arena (LMArena)

Human preference

Benchmarknot statedSourceTrend
Arena Text1445#86Arena (LMArena)
Arena Creative Writingnot in index1436#62Arena (LMArena)
Arena Multi-turnnot in index1466#70Arena (LMArena)
Arena Longer Queriesnot in index1443#99Arena (LMArena)
Lech Mazur Creative Writingnot in index75.6%#24Lech Mazur

Multimodal

Benchmarknot statedSourceTrend
Arena Vision1195#94Arena (LMArena)
VISTA42.1%#36Scale AI SEAL
MMMU (validation)74.4%#16MMMU
VPCTnot in index45.0%#13Epoch AI Benchmarking Hub

Long context

Benchmarknot statedSourceTrend
Fiction.LiveBench 120k63.9%#12Fiction.live

Composite

Benchmarknot statedSourceTrend
Epoch Capabilities Indexnot in index136.7#126Epoch AI Benchmarking Hub
AA Intelligence Index v4.3.29.6#395Artificial Analysis

Safety & honesty

Benchmarknot statedSourceTrend
MASKnot in index56.9%#41Scale AI SEAL

More from OpenAI

New this month

See also

Best for codingAll OpenAI models

Questions people ask

What are GPT-4.5's benchmark scores?
As of 25 Oct 2026, GPT-4.5 scores 49.7 ±4.9 on the BenchLeader Index, strongest in human preference (63) and weakest in reasoning (40). That rests on 19 independent benchmark results, and every score on this page links to its publisher.
How does GPT-4.5 rank?
GPT-4.5 is #388 of 785 ranked configurations on 25 Oct 2026. 323 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
When was GPT-4.5 released?
OpenAI released GPT-4.5 on 27 Feb 2025.

Data as of 25 Oct 2026. Compare with another model.

Cite as: BenchLeader, “GPT-4.5: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/gpt-4-5.