Noometry

Model comparison

DeepSeek LLM 67B vs Grok 4.5

Grok 4.5 is the stronger model overall, scoring 55.0 to 24.9 on the Noometry Index.

Last verified . 14 shared benchmarks.

DeepSeek LLM 67B DeepSeek

24.9

Rank #347 Confirmed

Grok 4.5 xAI

55.0

Rank #25 Confirmed

Summary

  • They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Grok 4.5 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Grok 4.5 leads 62.3 to 7.0.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 97.8% for Grok 4.5.
  • DeepSeek LLM 67B has downloadable open weights; the other is API-only.

Side by side

DeepSeek LLM 67BGrok 4.5
ProviderDeepSeekxAI
Noometry Index24.955.0
Released2023-11-292026-07-08
WeightsOpenProprietary
Context window—500K
Max output—500K
Input $ / M tokens—$2
Output $ / M tokens—$6
Results tracked1552

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

Category by category

Coding Grok 4.5 leads

DeepSeek LLM 67B: 31.9 (#278), Grok 4.5: 52.2 (#35)

BenchmarkDeepSeek LLM 67BGrok 4.5
LMArena Coding10961474
DeepSWE—53.8%
FrontierCode—42.4%
LMArena WebDev—1553
SciCode—54.1%
WeirdML—46.4%
ALE-Bench—1,309

Agentic & Tool Use Not comparable

DeepSeek LLM 67B: —, Grok 4.5: 44.4 (#17)

BenchmarkDeepSeek LLM 67BGrok 4.5
APEX-Agents—56.2%
τ²-bench Banking—47.9%
PostTrainBench—23.4%
GBAEval—65.4%
GDP.pdf—14%
LMArena Search—1213
Vending-Bench 2—3,887

Reasoning Grok 4.5 leads

DeepSeek LLM 67B: 16.5 (#304), Grok 4.5: 56.1 (#25)

BenchmarkDeepSeek LLM 67BGrok 4.5
Chess Puzzles0%36%
LMArena Hard Prompts10701462
Epoch Capabilities Index110.5153.92
ARC-AGI-2—52.6%
SimpleBench—70%
Kagi LLM Benchmark—83.5%
NYT Connections (extended)—79.9%
ARC-AGI-1—87.2%
CritPt—15.4%
DTBench—96.5%
LMCA—45.2%
Surface Evolver Bench—74.4%

Math Grok 4.5 leads

DeepSeek LLM 67B: 8.7 (#324), Grok 4.5: 60.9 (#35)

BenchmarkDeepSeek LLM 67BGrok 4.5
OTIS Mock AIME 2024-20250.8%97.8%
LMArena Math11081459
FrontierMath (Tiers 1-3)—57.2%
FrontierMath Tier 4—24.4%
ProofBench—31%
MATH Level 56.4%—

Knowledge Grok 4.5 leads

DeepSeek LLM 67B: 7.0 (#313), Grok 4.5: 62.3 (#24)

BenchmarkDeepSeek LLM 67BGrok 4.5
GPQA Diamond24.6%93.4%
SimpleQA Verified—48.3%
LMArena Expert—1466

Multimodal Not comparable

DeepSeek LLM 67B: —, Grok 4.5: 37.6 (#72)

BenchmarkDeepSeek LLM 67BGrok 4.5
LMArena Vision—1288
Blueprint-Bench 2—27.3%
Furniture Assembly—22.5%
LMArena Document—1452

Multilingual Grok 4.5 leads

DeepSeek LLM 67B: 29.4 (#267), Grok 4.5: 54.4 (#42)

BenchmarkDeepSeek LLM 67BGrok 4.5
LMArena Non-English10731440
LMArena Chinese11321496
LMArena French—1456
LMArena German—1446
LMArena Japanese—1428
LMArena Korean—1404
LMArena Russian—1448
LMArena Spanish—1450

Instruction Following Grok 4.5 leads

DeepSeek LLM 67B: 55.4 (#277), Grok 4.5: 76.0 (#48)

BenchmarkDeepSeek LLM 67BGrok 4.5
LMArena Instruction Following10791446

Long Context Grok 4.5 leads

DeepSeek LLM 67B: 33.1 (#265), Grok 4.5: 44.8 (#56)

BenchmarkDeepSeek LLM 67BGrok 4.5
LMArena Longer Query10921463

Writing & Preference Grok 4.5 leads

DeepSeek LLM 67B: 31.6 (#282), Grok 4.5: 65.8 (#42)

BenchmarkDeepSeek LLM 67BGrok 4.5
LMArena Text11051448
LMArena Creative Writing10671442
LMArena Multi-Turn10821456
EQ-Bench Creative Writing—1579

Frequently asked questions

Is DeepSeek LLM 67B better than Grok 4.5?

Grok 4.5 is the stronger model overall, scoring 55.0 to 24.9 on the Noometry Index.

Is DeepSeek LLM 67B or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 31.9 in the Noometry coding category.

How many benchmarks do DeepSeek LLM 67B and Grok 4.5 share?

14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Grok 4.5 has 52.

Related comparisons

Go deeper