Noometry

Model comparison

GPT-5.4 vs Mistral Small

GPT-5.4 is the stronger model overall, scoring 59.4 to 33.4 on the Noometry Index. Mistral Small costs 21× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Last verified . 27 shared benchmarks.

GPT-5.4 OpenAI

59.4

Rank #16 Confirmed

Mistral Small Mistral AI

33.4

Rank #243 Confirmed

Summary

  • They share 27 benchmarks with published results for both. GPT-5.4 scores higher in 10 categories and Mistral Small in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.4 leads 73.5 to 16.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 97.8% for GPT-5.4 and 5.8% for Mistral Small.
  • Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $2.50 / $15 for GPT-5.4.
  • GPT-5.4 accepts more context: 1.05M tokens versus 262K.
  • Mistral Small has downloadable open weights; the other is API-only.

Side by side

GPT-5.4Mistral Small
ProviderOpenAIMistral AI
Noometry Index59.433.4
Released2026-03-052024-02-26
WeightsProprietaryOpen
Context window1.05M262K
Max output128K256K
Input $ / M tokens$2.50$0.15
Output $ / M tokens$15$0.60
Results tracked6839

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

Category by category

Coding GPT-5.4 leads

GPT-5.4: 52.6 (#33), Mistral Small: 34.0 (#247)

BenchmarkGPT-5.4Mistral Small
SciCode56.6%26.5%
LMArena Coding14971362
ALE-Bench1,607497.62
SWE-bench Verified76.9%—
DeepSWE51.8%—
LMArena WebDev1465—
GSO31.4%—
WeirdML77.7%—
BigCodeBench Instruct—36.1%
LiveBench Coding—36.2%
MirrorCode15.6%—
BigCodeBench Complete—46.6%
AlgoTune1.85—

Agentic & Tool Use GPT-5.4 leads

GPT-5.4: 46.5 (#13), Mistral Small: 28.1 (#93)

BenchmarkGPT-5.4Mistral Small
Terminal-Bench81.8%—
APEX-Agents52.4%—
Berkeley Function Calling Leaderboard—37.1%
τ²-bench Banking39.4%—
DeepResearch Bench35.1%—
PostTrainBench19%—
GBAEval45.1%—
LMArena Search1197—
METR Time Horizons74.3%—
Vending-Bench 26,144—

Reasoning GPT-5.4 leads

GPT-5.4: 61.8 (#19), Mistral Small: 19.8 (#250)

BenchmarkGPT-5.4Mistral Small
Kagi LLM Benchmark63.8%37.8%
CritPt23.4%0%
LMArena Hard Prompts14851335
DTBench94.4%70.9%
LMCA52%20.6%
ARC-AGI-274%—
NYT Connections (extended)91.3%—
ARC-AGI-193.7%—
Chess Puzzles44%—
EnigmaEval16%—
Thematic Generalization80%—
EBR-Bench25.4%—
LiveBench Reasoning—44.8%
Mystery Game Puzzles37%—
LiveBench Data Analysis—53.7%
Epoch Capabilities Index156.81—
ForecastBench59.5—
LiveBench—44%

Math GPT-5.4 leads

GPT-5.4: 73.5 (#19), Mistral Small: 16.4 (#293)

BenchmarkGPT-5.4Mistral Small
OTIS Mock AIME 2024-202597.8%5.8%
LMArena Math14881341
FrontierMath (Tiers 1-3)78.6%—
FrontierMath Tier 449%—
MathArena Final-Answer Competitions83.1%—
ProofBench56%—
LiveBench Math—39.9%
MATH Level 5—46.8%
FrontierMath (Feb 2025 set)47.6%—
FrontierMath Tier 4 (v1)27.1%—

Knowledge GPT-5.4 leads

GPT-5.4: 65.3 (#14), Mistral Small: 31.0 (#222)

BenchmarkGPT-5.4Mistral Small
GPQA Diamond93.3%47.5%
Vectara Hallucination Rate7%5.1%
LMArena Expert15071291
Humanity's Last Exam36.2%—
SimpleQA Verified45.1%—
MMLU—68.7%

Multimodal GPT-5.4 leads

GPT-5.4: 43.7 (#20), Mistral Small: 33.5 (#96)

BenchmarkGPT-5.4Mistral Small
LMArena Vision13031142
Blueprint-Bench 227.1%—
Furniture Assembly37.5%—
LMArena Document1471—

Multilingual GPT-5.4 leads

GPT-5.4: 56.2 (#23), Mistral Small: 45.5 (#169)

BenchmarkGPT-5.4Mistral Small
LMArena Non-English14651315
LMArena Chinese15191340
LMArena French14931337
LMArena German14721340
LMArena Japanese14851275
LMArena Korean14481259
LMArena Russian14801324
LMArena Spanish14541346

Instruction Following GPT-5.4 leads

GPT-5.4: 77.1 (#27), Mistral Small: 66.4 (#209)

BenchmarkGPT-5.4Mistral Small
LMArena Instruction Following14691310
LiveBench Instruction Following—63.7%

Long Context GPT-5.4 leads

GPT-5.4: 50.3 (#8), Mistral Small: 40.4 (#156)

BenchmarkGPT-5.4Mistral Small
LMArena Longer Query14731327
CL-bench27.9%—
CL-bench Life21.7%—

Writing & Preference GPT-5.4 leads

GPT-5.4: 71.9 (#17), Mistral Small: 52.5 (#171)

BenchmarkGPT-5.4Mistral Small
LMArena Text14691338
LMArena Creative Writing14391305
LMArena Multi-Turn14821344
EQ-Bench Creative Writing1840—
EQ-Bench 41272—
LiveBench Language—30.5%

Frequently asked questions

Is GPT-5.4 better than Mistral Small?

GPT-5.4 is the stronger model overall, scoring 59.4 to 33.4 on the Noometry Index. Mistral Small costs 21× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Which is cheaper, GPT-5.4 or Mistral Small?

Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5.4 lists at $2.50 and $15.

Is GPT-5.4 or Mistral Small better for coding?

GPT-5.4 scores higher on coding benchmarks: 52.6 versus 34.0 in the Noometry coding category.

Which has the bigger context window?

GPT-5.4 does, with 1.05M tokens against 262K.

How many benchmarks do GPT-5.4 and Mistral Small share?

27 benchmarks have published results for both models. GPT-5.4 has 68 scored results on Noometry and Mistral Small has 39.

Related comparisons

Go deeper