Noometry

Model comparison

Grok 4.5 vs Tulu 3 (Tülu 3) 70B

Grok 4.5 is the stronger model overall, scoring 55.0 to 33.0 on the Noometry Index.

Last verified . 13 shared benchmarks.

Grok 4.5 xAI

55.0

Rank #25 Confirmed

Tulu 3 (Tülu 3) 70B Allen Institute for AI (Ai2)

33.0

Rank #251 Confirmed

Summary

  • They share 13 benchmarks with published results for both. Grok 4.5 scores higher in 8 categories and Tulu 3 (Tülu 3) 70B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Grok 4.5 leads 60.9 to 14.2.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 97.8% for Grok 4.5 and 4.4% for Tulu 3 (Tülu 3) 70B.
  • Tulu 3 (Tülu 3) 70B has downloadable open weights; the other is API-only.

Side by side

Grok 4.5Tulu 3 (Tülu 3) 70B
ProviderxAIAllen Institute for AI (Ai2)
Noometry Index55.033.0
Released2026-07-082024-11-21
WeightsProprietaryOpen
Context window500K—
Max output500K—
Input $ / M tokens$2—
Output $ / M tokens$6—
Results tracked5214

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

Category by category

Coding Grok 4.5 leads

Grok 4.5: 52.2 (#35), Tulu 3 (Tülu 3) 70B: 36.0 (#214)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Coding14741235
DeepSWE53.8%—
FrontierCode42.4%—
LMArena WebDev1553—
SciCode54.1%—
WeirdML46.4%—
ALE-Bench1,309—

Agentic & Tool Use Not comparable

Grok 4.5: 44.4 (#17), Tulu 3 (Tülu 3) 70B: —

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
APEX-Agents56.2%—
τ²-bench Banking47.9%—
PostTrainBench23.4%—
GBAEval65.4%—
GDP.pdf14%—
LMArena Search1213—
Vending-Bench 23,887—

Reasoning Grok 4.5 leads

Grok 4.5: 56.1 (#25), Tulu 3 (Tülu 3) 70B: 23.9 (#169)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Hard Prompts14621220
ARC-AGI-252.6%—
SimpleBench70%—
Kagi LLM Benchmark83.5%—
NYT Connections (extended)79.9%—
ARC-AGI-187.2%—
CritPt15.4%—
Chess Puzzles36%—
DTBench96.5%—
LMCA45.2%—
Surface Evolver Bench74.4%—
Epoch Capabilities Index153.92—

Math Grok 4.5 leads

Grok 4.5: 60.9 (#35), Tulu 3 (Tülu 3) 70B: 14.2 (#303)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
OTIS Mock AIME 2024-202597.8%4.4%
LMArena Math14591242
FrontierMath (Tiers 1-3)57.2%—
FrontierMath Tier 424.4%—
ProofBench31%—
MATH Level 5—42.7%

Knowledge Grok 4.5 leads

Grok 4.5: 62.3 (#24), Tulu 3 (Tülu 3) 70B: 25.0 (#264)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
GPQA Diamond93.4%46.3%
SimpleQA Verified48.3%—
LMArena Expert1466—

Multimodal Not comparable

Grok 4.5: 37.6 (#72), Tulu 3 (Tülu 3) 70B: —

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Vision1288—
Blueprint-Bench 227.3%—
Furniture Assembly22.5%—
LMArena Document1452—

Multilingual Grok 4.5 leads

Grok 4.5: 54.4 (#42), Tulu 3 (Tülu 3) 70B: 39.9 (#222)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Non-English14401236
LMArena Chinese14961249
LMArena Russian14481246
LMArena French1456—
LMArena German1446—
LMArena Japanese1428—
LMArena Korean1404—
LMArena Spanish1450—

Instruction Following Grok 4.5 leads

Grok 4.5: 76.0 (#48), Tulu 3 (Tülu 3) 70B: 64.8 (#227)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Instruction Following14461233

Long Context Grok 4.5 leads

Grok 4.5: 44.8 (#56), Tulu 3 (Tülu 3) 70B: 37.1 (#222)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Longer Query14631224

Writing & Preference Grok 4.5 leads

Grok 4.5: 65.8 (#42), Tulu 3 (Tülu 3) 70B: 45.6 (#223)

BenchmarkGrok 4.5Tulu 3 (Tülu 3) 70B
LMArena Text14481256
LMArena Creative Writing14421231
LMArena Multi-Turn14561252
EQ-Bench Creative Writing1579—

Frequently asked questions

Is Grok 4.5 better than Tulu 3 (Tülu 3) 70B?

Grok 4.5 is the stronger model overall, scoring 55.0 to 33.0 on the Noometry Index.

Is Grok 4.5 or Tulu 3 (Tülu 3) 70B better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 36.0 in the Noometry coding category.

How many benchmarks do Grok 4.5 and Tulu 3 (Tülu 3) 70B share?

13 benchmarks have published results for both models. Grok 4.5 has 52 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.

Related comparisons

Go deeper