GPT-4.5 Preview
GPT-4.5 Preview is behind the leaders in multimodal tasks and coding. Too few results yet to rate reasoning, agentic tasks, safety, long context, math, multilingual tasks, instruction following, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Reasoning, Agentic, Safety, Long Context, Math, Multilingual, Instruction Following or Factuality.
Price
No current price is tracked for this model. See the rate card
Evidence
33results on29benchmarks
- 19 independently verified
- 1 aggregator
- 13 vendor-reported
From 15 sources · latest Oct 8, 2026 · How verification works
GPT-4.5 Preview benchmark results
33 results on 29 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
38.9% behind the leader3 of 6 ranked benchmarks measured
- 75.20Oct 7, 2026
- 72.30Oct 7, 2026
- CharXiv (reasoning)55.40Oct 7, 2026CharXiv-R
Show 2 more multimodal resultsHide 2 multimodal results
- 1194.80Aug 25, 2026
- 1225Jun 17, 2026
43.7% behind the leader1 of 10 ranked benchmarks measured
- 38.00Oct 7, 2026
Show 2 more coding resultsHide 2 coding results
- LiveBench · Coding75.00Aug 23, 2026livebench_coding@2025-04-07
- LiveBench · Coding75.00Jun 17, 2026livebench_coding@2025-04-07
0 of 6 ranked benchmarks measured
- 34.50May 10, 2026
- 0.80May 10, 2026
- GPQA Diamond69.50Oct 7, 2026GPQA
0 of 3 ranked benchmarks measured
- LiveBench · Instruction Following72.33Aug 23, 2026livebench_instruction_following@2025-04-07
- LiveBench · Instruction Following72.33Jun 17, 2026livebench_instruction_following@2025-04-07
- 43.80Oct 7, 2026
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- livebench_language62.00Aug 23, 2026livebench_language@2025-04-07
- 1445Jul 23, 2026
- livebench_language62.00Jun 17, 2026livebench_language@2025-04-07
- 74.15May 19, 2026
- 92.00May 10, 2026
- 95.11May 10, 2026
Show 13 more resultsHide 13 results
- AA Intelligence9.58Oct 8, 2026aa_intelligence_index
- 44.90May 1, 2026
- 99.41May 10, 2026
- 10.30May 10, 2026
- 97.00Oct 7, 2026
- 95.81May 10, 2026
- 88.00Oct 7, 2026
- 88.20Aug 31, 2026
- 85.10Oct 7, 2026
- 100.00May 10, 2026
- 62.50Oct 7, 2026
- TAU-bench (airline)50.00Oct 7, 2026TAU-bench Airline
- TAU-bench (retail)68.40Oct 7, 2026TAU-bench Retail
GPT-4.5 Preview: common questions
Who makes GPT-4.5 Preview?
GPT-4.5 Preview is made by OpenAI.
When was GPT-4.5 Preview released?
GPT-4.5 Preview was released on Feb 27, 2025, according to Artificial Analysis.
What is GPT-4.5 Preview good at?
GPT-4.5 Preview is behind the leaders in multimodal tasks and coding. Too few results yet to rate reasoning, agentic tasks, safety, long context, math, multilingual tasks, instruction following, or factuality.
How many benchmarks has GPT-4.5 Preview been tested on?
We track 33 results for GPT-4.5 Preview on 29 benchmarks from 15 sources, 19 of them independently verified. The latest was recorded on Oct 8, 2026.
About this record
Where GPT-4.5 Preview's numbers come from, and every name it appears under.
- Tracked since
- May 1, 2026
- Newest source mention
- Aug 24, 2026
Where the results come from
Verification: 33 scores · 19 independently verified · 1 aggregator-attributed · 13 vendor-reported. How these tiers are assigned
From 15 sources on 9 sites. api.llm-stats.com supplies 13 of them; the 19 independently verified results come from 7 sites. Bars are coloured by trust tier.
- api.llm-stats.com13
- huggingface.co6
- storage.googleapis.com6
- arcprize.org2
- lmarena.ai2
- aider.chat1
- artificialanalysis.ai1
- datasets-server.huggingface.co1
- simple-bench.com1