Noometry

Model comparison

GPT-3.5-turbo vs Mistral Nemo

Mistral Nemo is the stronger model overall, scoring 25.9 to 23.1 on the Noometry Index.

Last verified . 7 shared benchmarks.

GPT-3.5-turbo OpenAI

23.1

Rank #345 Confirmed

Mistral Nemo Mistral AI

25.9

Rank #332 Confirmed

Summary

  • They share 7 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Mistral Nemo in 4 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in math, where Mistral Nemo leads 23.5 to 6.0.
  • The biggest single-benchmark swing is MATH Level 5: 15.9% for GPT-3.5-turbo and 10.8% for Mistral Nemo.
  • Mistral Nemo is cheaper at $0.15 / $0.15 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
  • Mistral Nemo accepts more context: 128K tokens versus 16K.
  • Mistral Nemo has downloadable open weights; the other is API-only.

Side by side

GPT-3.5-turboMistral Nemo
ProviderOpenAIMistral AI
Noometry Index23.125.9
Released2023-03-012024-07-01
WeightsProprietaryOpen
Context window16K128K
Max output4K128K
Input $ / M tokens$0.50$0.15
Output $ / M tokens$1.50$0.15
Results tracked4410

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

Category by category

Coding Not comparable

GPT-3.5-turbo: 23.5 (#326), Mistral Nemo: —

BenchmarkGPT-3.5-turboMistral Nemo
WeirdML3.5%—
BigCodeBench Instruct39.1%—
LMArena Coding1137—
BigCodeBench Complete50.6%—
HumanEval+70.7%—
MBPP+69.7%—

Agentic & Tool Use Not comparable

GPT-3.5-turbo: —, Mistral Nemo: 23.3 (#118)

BenchmarkGPT-3.5-turboMistral Nemo
Berkeley Function Calling Leaderboard—27.6%
BALROG—17.6%
METR Time Horizons21.5%—

Reasoning Mistral Nemo leads

GPT-3.5-turbo: 13.3 (#328), Mistral Nemo: 19.9 (#230)

BenchmarkGPT-3.5-turboMistral Nemo
DTBench48.5%48.6%
Epoch Capabilities Index118.48118.63
Chess Puzzles0%—
LMArena Hard Prompts1108—
Mystery Game Puzzles3%—
LMCA9.7%—
Adversarial NLI58.1%—
BIG-Bench Hard61.6%—
CommonsenseQA 2.057%—
ForecastBench50.4—
PIQA—83.5%
WinoGrande81.6%—

Math Mistral Nemo leads

GPT-3.5-turbo: 6.0 (#321), Mistral Nemo: 23.5 (#263)

BenchmarkGPT-3.5-turboMistral Nemo
MATH Level 515.9%10.8%
GSM8K57.8%84.2%
FrontierMath (Tiers 1-3)0%—
OTIS Mock AIME 2024-20252.2%—
LMArena Math1142—

Knowledge Mistral Nemo leads

GPT-3.5-turbo: 10.0 (#298), Mistral Nemo: 12.3 (#293)

BenchmarkGPT-3.5-turboMistral Nemo
GPQA Diamond28%29.9%
BoolQ87%82.5%
LMArena Expert1070—
ARC (AI2) Challenge87.4%—
MMLU71.4%—
OpenBookQA86%—
TriviaQA85.8%—

Multilingual Not comparable

GPT-3.5-turbo: 31.6 (#253), Mistral Nemo: —

BenchmarkGPT-3.5-turboMistral Nemo
LMArena Non-English1108—
LMArena Chinese1075—
LMArena French1117—
LMArena German1090—
LMArena Japanese1043—
LMArena Korean1019—
LMArena Russian1123—
LMArena Spanish1120—

Instruction Following Not comparable

GPT-3.5-turbo: 58.1 (#257), Mistral Nemo: —

BenchmarkGPT-3.5-turboMistral Nemo
LMArena Instruction Following1119—

Long Context Not comparable

GPT-3.5-turbo: 34.0 (#249), Mistral Nemo: —

BenchmarkGPT-3.5-turboMistral Nemo
LMArena Longer Query1121—

Writing & Preference Mistral Nemo leads

GPT-3.5-turbo: 25.4 (#300), Mistral Nemo: 28.6 (#291)

BenchmarkGPT-3.5-turboMistral Nemo
EQ-Bench Creative Writing451881
LMArena Text1125—
LMArena Creative Writing1092—
LMArena Multi-Turn1117—

Frequently asked questions

Is GPT-3.5-turbo better than Mistral Nemo?

Mistral Nemo is the stronger model overall, scoring 25.9 to 23.1 on the Noometry Index.

Which is cheaper, GPT-3.5-turbo or Mistral Nemo?

Mistral Nemo is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.

Which has the bigger context window?

Mistral Nemo does, with 128K tokens against 16K.

How many benchmarks do GPT-3.5-turbo and Mistral Nemo share?

7 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Mistral Nemo has 10.

Related comparisons

Go deeper