Noometry

Model comparison

DeepSeek LLM 67B vs GPT-5.4 nano

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 24.8 on the Noometry Index.

Last verified . 14 shared benchmarks.

DeepSeek LLM 67B DeepSeek

24.8

Rank #341 Confirmed

GPT-5.4 nano OpenAI

41.9

Rank #111 Confirmed

Summary

  • They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and GPT-5.4 nano in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-5.4 nano leads 41.8 to 7.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 87.8% for GPT-5.4 nano.
  • DeepSeek LLM 67B has downloadable open weights; the other is API-only.

Side by side

DeepSeek LLM 67BGPT-5.4 nano
ProviderDeepSeekOpenAI
Noometry Index24.841.9
Released2023-11-292026-03-17
WeightsOpenProprietary
Context window—400K
Max output—128K
Input $ / M tokens—$0.20
Output $ / M tokens—$1.25
Results tracked1540

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

Category by category

Coding GPT-5.4 nano leads

DeepSeek LLM 67B: 31.3 (#273), GPT-5.4 nano: 43.1 (#82)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
LMArena Coding10961404
SciCode—46.9%
WeirdML—49.2%
ALE-Bench—1,005

Reasoning GPT-5.4 nano leads

DeepSeek LLM 67B: 15.9 (#300), GPT-5.4 nano: 23.8 (#159)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
Chess Puzzles0%30%
LMArena Hard Prompts10701380
Epoch Capabilities Index110.54145.78
ARC-AGI-2—5.7%
Kagi LLM Benchmark—39.7%
ARC-AGI-1—51.5%
CritPt—9.3%
Mystery Game Puzzles—9%
DTBench—80.3%
LMCA—36.9%
ForecastBench—57.3

Math GPT-5.4 nano leads

DeepSeek LLM 67B: 8.1 (#318), GPT-5.4 nano: 40.7 (#76)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
OTIS Mock AIME 2024-20250.8%87.8%
LMArena Math11081406
FrontierMath (Tiers 1-3)—44.9%
FrontierMath Tier 4—12.2%
ProofBench—5%
MATH Level 56.4%—
FrontierMath (Feb 2025 set)—25.9%
FrontierMath Tier 4 (v1)—6.3%

Knowledge GPT-5.4 nano leads

DeepSeek LLM 67B: 7.1 (#308), GPT-5.4 nano: 41.8 (#94)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
GPQA Diamond24.6%78.5%
SimpleQA Verified—11.7%
Vectara Hallucination Rate—3.1%
LMArena Expert—1397

Multimodal Not comparable

DeepSeek LLM 67B: —, GPT-5.4 nano: 36.0 (#73)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
LMArena Vision—1196

Multilingual GPT-5.4 nano leads

DeepSeek LLM 67B: 29.5 (#262), GPT-5.4 nano: 48.8 (#136)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
LMArena Non-English10731359
LMArena Chinese11321391
LMArena French—1396
LMArena German—1368
LMArena Japanese—1342
LMArena Korean—1321
LMArena Russian—1363
LMArena Spanish—1368

Instruction Following GPT-5.4 nano leads

DeepSeek LLM 67B: 55.7 (#272), GPT-5.4 nano: 72.1 (#139)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
LMArena Instruction Following10791362

Long Context GPT-5.4 nano leads

DeepSeek LLM 67B: 33.1 (#260), GPT-5.4 nano: 41.6 (#131)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
LMArena Longer Query10921365

Writing & Preference GPT-5.4 nano leads

DeepSeek LLM 67B: 31.8 (#277), GPT-5.4 nano: 55.9 (#137)

BenchmarkDeepSeek LLM 67BGPT-5.4 nano
LMArena Text11051372
LMArena Creative Writing10671313
LMArena Multi-Turn10821382

Frequently asked questions

Is DeepSeek LLM 67B better than GPT-5.4 nano?

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 24.8 on the Noometry Index.

Is DeepSeek LLM 67B or GPT-5.4 nano better for coding?

GPT-5.4 nano scores higher on coding benchmarks: 43.1 versus 31.3 in the Noometry coding category.

How many benchmarks do DeepSeek LLM 67B and GPT-5.4 nano share?

14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and GPT-5.4 nano has 40.

Related comparisons

Go deeper