Noometry

Model comparison

GPT-4 vs Magistral Medium

Magistral Medium is the stronger model overall, scoring 35.2 to 29.1 on the Noometry Index.

Last verified . 17 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Magistral Medium Mistral AI

35.2

Rank #227 Confirmed

Summary

  • They share 17 benchmarks with published results for both. GPT-4 scores higher in 2 categories and Magistral Medium in 6 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in math, where Magistral Medium leads 35.1 to 10.8.
  • Magistral Medium is cheaper at $2 / $5 per million input/output tokens, against $30 / $60 for GPT-4.
  • Magistral Medium accepts more context: 262K tokens versus 8K.
  • Magistral Medium has downloadable open weights; the other is API-only.

Side by side

GPT-4Magistral Medium
ProviderOpenAIMistral AI
Noometry Index29.135.2
Released2023-03-142025-03-17
WeightsProprietaryOpen
Context window8K262K
Max output8K16K
Input $ / M tokens$30$2
Output $ / M tokens$60$5
Results tracked3822

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

Category by category

Coding Magistral Medium leads

GPT-4: 31.6 (#283), Magistral Medium: 39.1 (#161)

BenchmarkGPT-4Magistral Medium
LMArena Coding12541319
SciCode—39.2%
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Magistral Medium: —

BenchmarkGPT-4Magistral Medium
METR Time Horizons36.1%—

Reasoning GPT-4 leads

GPT-4: 17.8 (#289), Magistral Medium: 8.6 (#348)

BenchmarkGPT-4Magistral Medium
LMArena Hard Prompts12411267
ARC-AGI-2—0%
Kagi LLM Benchmark—16.2%
ARC-AGI-1—6.1%
CritPt—0.3%
Chess Puzzles4%—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.89—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Magistral Medium leads

GPT-4: 10.8 (#309), Magistral Medium: 35.1 (#189)

BenchmarkGPT-4Magistral Medium
LMArena Math12691250
OTIS Mock AIME 2024-20251.1%—
MATH Level 523%—
GSM8K92%—

Knowledge Magistral Medium leads

GPT-4: 18.4 (#282), Magistral Medium: 33.5 (#202)

BenchmarkGPT-4Magistral Medium
LMArena Expert12111223
GPQA Diamond35.7%—
MMLU86.4%—
TriviaQA84.8%—

Multilingual Too close to call

GPT-4: 40.6 (#215), Magistral Medium: 39.6 (#224)

BenchmarkGPT-4Magistral Medium
LMArena Non-English12461232
LMArena Chinese12421227
LMArena French12831267
LMArena German12511248
LMArena Japanese12091175
LMArena Korean11841125
LMArena Russian12511224
LMArena Spanish12611271

Instruction Following Too close to call

GPT-4: 65.3 (#222), Magistral Medium: 66.0 (#211)

BenchmarkGPT-4Magistral Medium
LMArena Instruction Following12411254

Long Context Magistral Medium leads

GPT-4: 37.7 (#212), Magistral Medium: 39.3 (#183)

BenchmarkGPT-4Magistral Medium
LMArena Longer Query12441295

Writing & Preference Magistral Medium leads

GPT-4: 34.9 (#268), Magistral Medium: 46.3 (#219)

BenchmarkGPT-4Magistral Medium
LMArena Text12631255
LMArena Creative Writing12441245
LMArena Multi-Turn12571275
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than Magistral Medium?

Magistral Medium is the stronger model overall, scoring 35.2 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 or Magistral Medium?

Magistral Medium is cheaper. It lists at $2 per million input tokens and $5 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or Magistral Medium better for coding?

Magistral Medium scores higher on coding benchmarks: 39.1 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

Magistral Medium does, with 262K tokens against 8K.

How many benchmarks do GPT-4 and Magistral Medium share?

17 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Magistral Medium has 22.

Related comparisons

Go deeper