Noometry

Model comparison

GPT-4 vs Mistral Nemo

GPT-4 is the stronger model overall, scoring 29.1 to 26.4 on the Noometry Index. Mistral Nemo costs 250× less per token, which makes it the better buy when GPT-4's lead doesn't matter for your workload.

Last verified . 6 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Mistral Nemo Mistral AI

26.4

Rank #337 Confirmed

Summary

  • They share 6 benchmarks with published results for both. GPT-4 scores higher in 2 categories and Mistral Nemo in 2 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in math, where Mistral Nemo leads 25.5 to 10.8.
  • The biggest single-benchmark swing is DTBench: 62.7% for GPT-4 and 48.6% for Mistral Nemo.
  • Mistral Nemo is cheaper at $0.15 / $0.15 per million input/output tokens, against $30 / $60 for GPT-4.
  • Mistral Nemo accepts more context: 128K tokens versus 8K.
  • Mistral Nemo has downloadable open weights; the other is API-only.

Side by side

GPT-4Mistral Nemo
ProviderOpenAIMistral AI
Noometry Index29.126.4
Released2023-03-142024-07-01
WeightsProprietaryOpen
Context window8K128K
Max output8K128K
Input $ / M tokens$30$0.15
Output $ / M tokens$60$0.15
Results tracked3810

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

Category by category

Coding Not comparable

GPT-4: 31.6 (#283), Mistral Nemo: —

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

Agentic & Tool Use Not comparable

GPT-4: —, Mistral Nemo: 23.5 (#125)

BenchmarkGPT-4Mistral Nemo
Berkeley Function Calling Leaderboard—27.6%
BALROG—17.6%
METR Time Horizons36.1%—

Reasoning Mistral Nemo leads

GPT-4: 17.8 (#289), Mistral Nemo: 20.7 (#232)

BenchmarkGPT-4Mistral Nemo
DTBench62.7%48.6%
Epoch Capabilities Index125.89118.68
Chess Puzzles4%—
LMArena Hard Prompts1241—
Mystery Game Puzzles12%—
LMCA17.1%—
BIG-Bench Hard75.1%—
ForecastBench57.8—
HellaSwag95.3%—
PIQA—83.5%
WinoGrande87.5%—

Math Mistral Nemo leads

GPT-4: 10.8 (#309), Mistral Nemo: 25.5 (#268)

BenchmarkGPT-4Mistral Nemo
MATH Level 523%10.8%
GSM8K92%84.2%
OTIS Mock AIME 2024-20251.1%—
LMArena Math1269—

Knowledge GPT-4 leads

GPT-4: 18.4 (#282), Mistral Nemo: 12.3 (#298)

BenchmarkGPT-4Mistral Nemo
GPQA Diamond35.7%29.9%
LMArena Expert1211—
BoolQ—82.5%
MMLU86.4%—
TriviaQA84.8%—

Multilingual Not comparable

GPT-4: 40.6 (#215), Mistral Nemo: —

BenchmarkGPT-4Mistral Nemo
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), Mistral Nemo: —

BenchmarkGPT-4Mistral Nemo
LMArena Instruction Following1241—

Long Context Not comparable

GPT-4: 37.7 (#212), Mistral Nemo: —

BenchmarkGPT-4Mistral Nemo
LMArena Longer Query1244—

Writing & Preference GPT-4 leads

GPT-4: 34.9 (#268), Mistral Nemo: 28.5 (#296)

BenchmarkGPT-4Mistral Nemo
EQ-Bench Creative Writing752881
LMArena Text1263—
LMArena Creative Writing1244—
LMArena Multi-Turn1257—

Frequently asked questions

Is GPT-4 better than Mistral Nemo?

GPT-4 is the stronger model overall, scoring 29.1 to 26.4 on the Noometry Index. Mistral Nemo costs 250× less per token, which makes it the better buy when GPT-4's lead doesn't matter for your workload.

Which is cheaper, GPT-4 or Mistral Nemo?

Mistral Nemo is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; GPT-4 lists at $30 and $60.

Which has the bigger context window?

Mistral Nemo does, with 128K tokens against 8K.

How many benchmarks do GPT-4 and Mistral Nemo share?

6 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Mistral Nemo has 10.

Related comparisons

Go deeper