Inference
EN
OpenAI

GPT-4.5

GPT-4.5 is a standard model from OpenAI. 9 benchmarks count toward its index, in 5 categories.

  • Closed weights
  • Standard
  • Input: text
  • Released 2025-02
RankedEnough results for an index.
IndexAll categories, out of 100.38.8±8.9Median 51.0
CoverageIndex weight that has results.65%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.$75.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$150.00Median $2.55
ContextMaximum tokens in one request.128KMedian 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.None counted——
CodingCode writing and repair.1 counted29.1—
ReasoningLogic problems and puzzles.2 counted35.6188 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.3 counted48.8144 / 268
MultilingualTasks in many languages.1 counted31.8—
InstructionTasks with strict prompt rules.None counted——
MathMath problems.2 counted34.0138 / 168

Results

9 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Massive Multitask Language UnderstandingKnowledge90.8%2 / 34
Measuring Short-Form Factuality in Large Language ModelsKnowledgeAbove fitFar above this model's index.62.5%1 / 14
Multilingual Grade School MathMultilingual86.9%4 / 13
Evaluating Large Language Models Trained on CodeCoding88.6%7 / 22
MATH level 5Math78.6% ±1.129 / 88
GPQA diamond · Epoch AIKnowledge68.7% ±3.3115 / 217
OTIS Mock AIME 2024-2025Math37.8% ±7.3129 / 200
Graduate-Level Google-Proof Q&AKnowledge69.5%94 / 141
ARC-AGI-2 (semi-private)Reasoning0.8%113 / 131
ARC-AGI-1 (semi-private)Reasoning10.3%115 / 126
ARC-AGI-2 (public eval)Reasoning0.0%102 / 122
Discrete Reasoning Over ParagraphsReasoning83.4%—

Sources

BenchLMCC BY-NC 4.0Epoch AICC BYARC Prize FoundationTerms restrict reuse

History

5 releases
Index39.8 → 38.8
23 Oct 202612 Nov 2026
Output / 1M$150.00 → $150.00
23 Oct 202612 Nov 2026

Questions

How much does GPT-4.5 cost?
$75.00 per 1M input tokens, $150.00 per 1M output tokens.
How good is GPT-4.5?
Its index is 38.8 out of 100. That is place 216 of 283 ranked models.
What is GPT-4.5 best at?
Knowledge: 48.8 out of 100, place 144 of 268.
What is the context window of GPT-4.5?
128K tokens.
Is GPT-4.5 open source?
No. Its weights are not published.
When was GPT-4.5 released?
February 2025.

Badge

GPT-4.5 on Inference360: place 216 of 283, index 38.8
HTML
<a href="https://inference360.ai/models/gpt-4-5"><img src="https://inference360.ai/badges/gpt-4-5.svg" alt="GPT-4.5 on Inference360: place 216 of 283, index 38.8" height="20"></a>
Markdown
[![GPT-4.5 on Inference360: place 216 of 283, index 38.8](https://inference360.ai/badges/gpt-4-5.svg)](https://inference360.ai/models/gpt-4-5)

Compare with

DeepSeek-R138.9GPT-4.138.0o3-mini40.0

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from OpenAI

All models
o3 Mini HighUnrankedo1-proUnrankedo3-mini40.0GPT-4.138.0GPT-4.1 mini34.5