Noometry

Model comparison

GPT-5.3 Chat vs GPT-5.4 mini

GPT-5.4 mini is the stronger model overall, scoring 44.8 to 42.1 on the Noometry Index.

Last verified . 18 shared benchmarks.

GPT-5.3 Chat OpenAI

42.1

Rank #109 Confirmed

GPT-5.4 mini OpenAI

44.8

Rank #69 Confirmed

Summary

  • They share 18 benchmarks with published results for both. GPT-5.3 Chat scores higher in 0 categories and GPT-5.4 mini in 8 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-5.4 mini leads 51.4 to 38.2.
  • GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Chat.
  • GPT-5.4 mini accepts more context: 400K tokens versus 128K.

Side by side

GPT-5.3 ChatGPT-5.4 mini
ProviderOpenAIOpenAI
Noometry Index42.144.8
Released2026-03-032026-03-17
WeightsProprietaryProprietary
Context window128K400K
Max output16K128K
Input $ / M tokens$1.75$0.75
Output $ / M tokens$14$4.50
Results tracked1846

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

Category by category

Coding GPT-5.4 mini leads

GPT-5.3 Chat: 40.8 (#124), GPT-5.4 mini: 45.0 (#67)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Coding14081438
FrontierCode—27%
LMArena WebDev—1397
SciCode—49.9%
WeirdML—60.3%
ALE-Bench—1,189

Agentic & Tool Use Not comparable

GPT-5.3 Chat: —, GPT-5.4 mini: 29.6 (#74)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
DeepResearch Bench—36.3%

Reasoning GPT-5.4 mini leads

GPT-5.3 Chat: 27.3 (#105), GPT-5.4 mini: 30.3 (#77)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Hard Prompts13991424
ARC-AGI-2—18.9%
Kagi LLM Benchmark—37.9%
NYT Connections (extended)—61.8%
ARC-AGI-1—63.7%
CritPt—10%
Chess Puzzles—24%
Thematic Generalization—61.7%
Mystery Game Puzzles—11%
DTBench—80%
LMCA—40.8%
Epoch Capabilities Index—148.83
ForecastBench—57

Math GPT-5.4 mini leads

GPT-5.3 Chat: 35.4 (#146), GPT-5.4 mini: 45.4 (#63)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Math13891418
FrontierMath (Tiers 1-3)—51.2%
FrontierMath Tier 4—9.8%
OTIS Mock AIME 2024-2025—88.9%
ProofBench—21%
FrontierMath (Feb 2025 set)—28.3%
FrontierMath Tier 4 (v1)—2.1%

Knowledge GPT-5.4 mini leads

GPT-5.3 Chat: 38.2 (#137), GPT-5.4 mini: 51.4 (#62)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Expert13971434
GPQA Diamond—86.9%
SimpleQA Verified—29.4%
Vectara Hallucination Rate—5.5%

Multimodal Not comparable

GPT-5.3 Chat: —, GPT-5.4 mini: 39.0 (#51)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Vision—1245

Multilingual GPT-5.4 mini leads

GPT-5.3 Chat: 50.5 (#119), GPT-5.4 mini: 52.1 (#94)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Non-English13831405
LMArena Chinese14331446
LMArena French13981440
LMArena German13841407
LMArena Japanese13511375
LMArena Korean13471369
LMArena Russian14001417
LMArena Spanish13711403

Instruction Following GPT-5.4 mini leads

GPT-5.3 Chat: 73.0 (#124), GPT-5.4 mini: 74.3 (#99)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Instruction Following13781405

Long Context Too close to call

GPT-5.3 Chat: 42.6 (#115), GPT-5.4 mini: 42.9 (#107)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Longer Query13961407

Writing & Preference Too close to call

GPT-5.3 Chat: 63.3 (#65), GPT-5.4 mini: 64.2 (#56)

BenchmarkGPT-5.3 ChatGPT-5.4 mini
LMArena Text13891411
LMArena Creative Writing13561370
EQ-Bench Creative Writing16901665
LMArena Multi-Turn14131429

Frequently asked questions

Is GPT-5.3 Chat better than GPT-5.4 mini?

GPT-5.4 mini is the stronger model overall, scoring 44.8 to 42.1 on the Noometry Index.

Which is cheaper, GPT-5.3 Chat or GPT-5.4 mini?

GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; GPT-5.3 Chat lists at $1.75 and $14.

Is GPT-5.3 Chat or GPT-5.4 mini better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 40.8 in the Noometry coding category.

Which has the bigger context window?

GPT-5.4 mini does, with 400K tokens against 128K.

How many benchmarks do GPT-5.3 Chat and GPT-5.4 mini share?

18 benchmarks have published results for both models. GPT-5.3 Chat has 18 scored results on Noometry and GPT-5.4 mini has 46.

Related comparisons

Go deeper