Noometry

Model comparison

GPT-4 vs Mistral Small 3

Mistral Small 3 is the stronger model overall, scoring 30.9 to 28.9 on the Noometry Index.

Last verified . 23 shared benchmarks.

GPT-4 OpenAI

28.9

Rank #308 Confirmed

Mistral Small 3 Mistral AI

30.9

Rank #268 Confirmed

Summary

  • They share 23 benchmarks with published results for both. GPT-4 scores higher in 3 categories and Mistral Small 3 in 5 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Mistral Small 3 leads 25.0 to 18.4.
  • The biggest single-benchmark swing is GPQA Diamond: 35.7% for GPT-4 and 47.3% for Mistral Small 3.
  • Mistral Small 3 is cheaper at $0.05 / $0.08 per million input/output tokens, against $30 / $60 for GPT-4.
  • Mistral Small 3 accepts more context: 33K tokens versus 8K.
  • Mistral Small 3 has downloadable open weights; the other is API-only.

Side by side

GPT-4Mistral Small 3
ProviderOpenAIMistral AI
Noometry Index28.930.9
Released2023-03-142025-01-30
WeightsProprietaryOpen
Context window8K33K
Max output8K16K
Input $ / M tokens$30$0.05
Output $ / M tokens$60$0.08
Results tracked3824

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

Category by category

Coding Mistral Small 3 leads

GPT-4: 31.2 (#277), Mistral Small 3: 36.0 (#203)

BenchmarkGPT-4Mistral Small 3
BigCodeBench Instruct46%45.3%
LMArena Coding12541246
BigCodeBench Complete57.2%50.4%
WeirdML12.4%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Mistral Small 3: —

BenchmarkGPT-4Mistral Small 3
METR Time Horizons36.1%—

Reasoning Too close to call

GPT-4: 17.3 (#285), Mistral Small 3: 18.2 (#269)

BenchmarkGPT-4Mistral Small 3
Chess Puzzles4%0%
LMArena Hard Prompts12411233
Epoch Capabilities Index125.88127.06
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Mistral Small 3 leads

GPT-4: 10.1 (#304), Mistral Small 3: 15.3 (#290)

BenchmarkGPT-4Mistral Small 3
OTIS Mock AIME 2024-20251.1%6.7%
LMArena Math12691240
MATH Level 523%—
GSM8K92%—

Knowledge Mistral Small 3 leads

GPT-4: 18.4 (#277), Mistral Small 3: 25.0 (#258)

BenchmarkGPT-4Mistral Small 3
GPQA Diamond35.7%47.3%
LMArena Expert12111202
Confabulations—25.2%
MMLU86.4%—
TriviaQA84.8%—

Multilingual GPT-4 leads

GPT-4: 40.8 (#210), Mistral Small 3: 37.5 (#231)

BenchmarkGPT-4Mistral Small 3
LMArena Non-English12461198
LMArena Chinese12421204
LMArena French12821202
LMArena German12511211
LMArena Japanese12091111
LMArena Korean11841188
LMArena Russian12511216
LMArena Spanish1261—

Instruction Following GPT-4 leads

GPT-4: 65.5 (#217), Mistral Small 3: 63.9 (#224)

BenchmarkGPT-4Mistral Small 3
LMArena Instruction Following12411214

Long Context Too close to call

GPT-4: 37.7 (#207), Mistral Small 3: 37.8 (#206)

BenchmarkGPT-4Mistral Small 3
LMArena Longer Query12441246

Writing & Preference GPT-4 leads

GPT-4: 35.0 (#263), Mistral Small 3: 32.4 (#275)

BenchmarkGPT-4Mistral Small 3
LMArena Text12631234
LMArena Creative Writing12441195
EQ-Bench Creative Writing752707
LMArena Multi-Turn12571217

Frequently asked questions

Is GPT-4 better than Mistral Small 3?

Mistral Small 3 is the stronger model overall, scoring 30.9 to 28.9 on the Noometry Index.

Which is cheaper, GPT-4 or Mistral Small 3?

Mistral Small 3 is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or Mistral Small 3 better for coding?

Mistral Small 3 scores higher on coding benchmarks: 36.0 versus 31.2 in the Noometry coding category.

Which has the bigger context window?

Mistral Small 3 does, with 33K tokens against 8K.

How many benchmarks do GPT-4 and Mistral Small 3 share?

23 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Mistral Small 3 has 24.

Related comparisons

Go deeper