Noometry

OpenAI, proprietary

GPT-4.5

GPT-4.5 by OpenAI ranks 208th of 354 ranked models on the Noometry Index as of October 2026, with a score of 37.2. Its strongest category is multimodal, where it ranks 71st.

Last verified

Noometry rank
#208 of 354
Index score
37.2
Evidence
Confirmed 42 results
Provider
OpenAI
Released
February 27, 2025
Weights
Proprietary
Reasoning
Unknown
Context window
—
Max output
—
Input price
Not listed
Output price
Not listed
Blended price
Not listed
Output speed
Not measured
Value
Not ranked
Knowledge cutoff
Unknown

Category scores

Each category score combines every public result we have in that category.

  1. Coding 42.2
  2. Agentic & Tool Use 27.9
  3. Reasoning 13.9
  4. Math 32.6
  5. Knowledge 32.5
  6. Multimodal 37.6
  7. Multilingual 52.5
  8. Instruction Following 72.6
  9. Long Context 40.4
  10. Writing & Preference 56.9
CategoryScoreRankResults
Coding42.2#1094
Agentic & Tool Use27.9#971
Reasoning13.9#3307
Math32.6#2114
Knowledge32.5#2114
Multimodal37.6#712
Multilingual52.5#831
Instruction Following72.6#1342
Long Context40.4#1552
Writing & Preference56.9#1346

Strengths and weaknesses

Categories where GPT-4.5 places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Multilingual52.5+5.1#83 of 297, top 28%
Coding42.2+3.5#109 of 340, top 33%
Writing & Preference56.9+3.1#134 of 312, top 43%

Weakest categories

CategoryScorevs medianRank
Reasoning13.9−9.7#330 of 350, top 95%
Knowledge32.5−4.8#211 of 314, top 68%
Math32.6−3.9#211 of 327, top 65%

Closest competitors

The models ranked just above and below GPT-4.5. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
MiniMax-M2#20437.4$0.5217Compare
Granite 4.1 8b#20537.4——Compare
Gemma 3n E4b IT#20637.3—13Compare
Step 3.7 Flash#20737.3$0.42—Compare
Yi-Lightning#20937.1——Compare
Qwen Plus#21037.1$0.6037Compare
Gemini 2.5 Flash-Lite#21137.0$0.18172Compare
o3-mini#21236.7$1.93—Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
Aider Polyglot44.9%#25 of 44, top 57%Epoch AI
WeirdML39.4%#80 of 119, top 68%Epoch AI
LiveBench Coding75.2%#3 of 39, top 8%Epoch AI
LMArena Coding1396#139 of 294, top 48%LMArena2026-10-08

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
Cybench17.5%#13 of 21, top 62%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
ARC-AGI-20.8%#72 of 83, top 87%Epoch AI
SimpleBench34.5%#58 of 77, top 76%Epoch AI
ARC-AGI-110.3%#74 of 83, top 90%Epoch AI
EnigmaEval3.2%#26 of 38, top 69%Epoch AI
LiveBench Reasoning71.1%#10 of 39, top 26%Epoch AI
LMArena Hard Prompts1403#122 of 297, top 42%LMArena2026-10-08
LiveBench Data Analysis64.3%#12 of 39, top 31%Epoch AI
Epoch Capabilities Index136.74#120 of 213, top 57%Epoch AI2025-02-27
ForecastBench61.7#7 of 72, top 10%Epoch AI
LiveBench69%#8 of 39, top 21%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
OTIS Mock AIME 2024-202537.8%#118 of 173, top 69%Epoch AI2025-02-28
LiveBench Math69.3%#11 of 39, top 29%Epoch AI
LMArena Math1412#112 of 285, top 40%LMArena2026-10-08
MATH Level 578.6%#26 of 79, top 33%Epoch AI2025-02-28

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond68.7%#100 of 186, top 54%Epoch AI2025-02-28
Humanity's Last Exam5.4%#34 of 41, top 83%Epoch AI
Confabulations (lower is better)13.6%#13 of 51, top 26%Lech Mazur benchmarks
LMArena Expert1394#130 of 273, top 48%LMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1195#81 of 122, top 67%LMArena2026-10-09
VPCT45%#10 of 24, top 42%Epoch AI

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1413#83 of 297, top 28%LMArena2026-10-08
LMArena Chinese1421#127 of 285, top 45%LMArena2026-10-08
LMArena French1418#101 of 223, top 46%LMArena2026-10-08
LMArena German1457#38 of 231, top 17%LMArena2026-10-08
LMArena Japanese1416#44 of 211, top 21%LMArena2026-10-08
LMArena Korean1392#65 of 213, top 31%LMArena2026-10-08
LMArena Russian1419#82 of 283, top 29%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LiveBench Instruction Following72.3%#16 of 39, top 42%Epoch AI
LMArena Instruction Following1404#100 of 298, top 34%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
Fiction.LiveBench63.9%#25 of 47, top 54%Epoch AI
LMArena Longer Query1406#112 of 291, top 39%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1417#99 of 297, top 34%LMArena2026-10-08
LMArena Creative Writing1394#89 of 295, top 31%LMArena2026-10-08
Short-Story Creative Writing75.6%#23 of 39, top 59%Epoch AI
EQ-Bench Creative Writing1258#82 of 115, top 72%EQ-Bench
LMArena Multi-Turn1444#68 of 295, top 24%LMArena2026-10-08
LiveBench Language61.5%#4 of 39, top 11%Epoch AI

Compare GPT-4.5

Other OpenAI models

Frequently asked questions

How good is GPT-4.5?

GPT-4.5 by OpenAI ranks 208th of 354 ranked models on the Noometry Index as of October 2026, with a score of 37.2. Its strongest category is multimodal, where it ranks 71st.

Is GPT-4.5 open source?

No. GPT-4.5 is proprietary and available only through OpenAI's API and partner platforms.

What are GPT-4.5's strengths and weaknesses?

Relative to other ranked models, GPT-4.5 places best in multilingual, coding, writing & preference and lowest in reasoning, knowledge, math.

What is GPT-4.5 best at?

Its best category is multimodal, where it ranks 71st on Noometry.