Noometry

Model comparison

GPT-4 vs Pixtral Large

Pixtral Large is the stronger model overall, scoring 32.2 to 29.1 on the Noometry Index.

Last verified . 1 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Pixtral Large Mistral AI

32.2

Rank #259 Reported

Summary

  • They share 1 benchmark with published results for both. GPT-4 scores higher in 1 category and Pixtral Large in 1 category; 2 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Pixtral Large leads 21.7 to 17.8.
  • Pixtral Large is cheaper at $2 / $6 per million input/output tokens, against $30 / $60 for GPT-4.
  • Pixtral Large accepts more context: 128K tokens versus 8K.
  • Pixtral Large has downloadable open weights; the other is API-only.

Side by side

GPT-4Pixtral Large
ProviderOpenAIMistral AI
Noometry Index29.132.2
Released2023-03-142024-11-01
WeightsProprietaryOpen
Context window8K128K
Max output8K128K
Input $ / M tokens$30$2
Output $ / M tokens$60$6
Results tracked383

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

Category by category

Coding Not comparable

GPT-4: 31.6 (#283), Pixtral Large: —

BenchmarkGPT-4Pixtral Large
WeirdML12.4%—
BigCodeBench Instruct46%—
LMArena Coding1254—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Pixtral Large: —

BenchmarkGPT-4Pixtral Large
METR Time Horizons36.1%—

Reasoning Pixtral Large leads

GPT-4: 17.8 (#289), Pixtral Large: 21.7 (#218)

BenchmarkGPT-4Pixtral Large
Chess Puzzles4%—
EnigmaEval—0.8%
LMArena Hard Prompts1241—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.89—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Not comparable

GPT-4: 10.8 (#309), Pixtral Large: —

BenchmarkGPT-4Pixtral Large
OTIS Mock AIME 2024-20251.1%—
LMArena Math1269—
MATH Level 523%—
GSM8K92%—

Knowledge Not comparable

GPT-4: 18.4 (#282), Pixtral Large: —

BenchmarkGPT-4Pixtral Large
GPQA Diamond35.7%—
LMArena Expert1211—
MMLU86.4%—
TriviaQA84.8%—

Multimodal Not comparable

GPT-4: —, Pixtral Large: 30.6 (#111)

BenchmarkGPT-4Pixtral Large
LMArena Vision—1089

Multilingual Not comparable

GPT-4: 40.6 (#215), Pixtral Large: —

BenchmarkGPT-4Pixtral Large
LMArena Non-English1246—
LMArena Chinese1242—
LMArena French1283—
LMArena German1251—
LMArena Japanese1209—
LMArena Korean1184—
LMArena Russian1251—
LMArena Spanish1261—

Instruction Following Not comparable

GPT-4: 65.3 (#222), Pixtral Large: —

BenchmarkGPT-4Pixtral Large
LMArena Instruction Following1241—

Long Context Not comparable

GPT-4: 37.7 (#212), Pixtral Large: —

BenchmarkGPT-4Pixtral Large
LMArena Longer Query1244—

Writing & Preference GPT-4 leads

GPT-4: 34.9 (#268), Pixtral Large: 32.9 (#278)

BenchmarkGPT-4Pixtral Large
EQ-Bench Creative Writing752988
LMArena Text1263—
LMArena Creative Writing1244—
LMArena Multi-Turn1257—

Frequently asked questions

Is GPT-4 better than Pixtral Large?

Pixtral Large is the stronger model overall, scoring 32.2 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 or Pixtral Large?

Pixtral Large is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-4 lists at $30 and $60.

Which has the bigger context window?

Pixtral Large does, with 128K tokens against 8K.

How many benchmarks do GPT-4 and Pixtral Large share?

1 benchmark has published results for both models. GPT-4 has 38 scored results on Noometry and Pixtral Large has 3.

Related comparisons

Go deeper