Noometry

Model comparison

Gemma 3 27B vs Tulu 3 (Tülu 3) 70B

Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 32.6 to 30.5 on the Noometry Index.

Last verified . 14 shared benchmarks.

Gemma 3 27B Google

30.5

Rank #277 Confirmed

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

32.6

Rank #242 Confirmed

Summary

  • They share 14 benchmarks with published results for both. Gemma 3 27B scores higher in 5 categories and Tulu 3 (Tülu 3) 70B in 3 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Tulu 3 (Tülu 3) 70B leads 35.4 to 22.1.
  • The biggest single-benchmark swing is MATH Level 5: 74% for Gemma 3 27B and 42.7% for Tulu 3 (Tülu 3) 70B.

Side by side

Gemma 3 27BTulu 3 (Tülu 3) 70B
ProviderGoogleAllen Institute for AI (Ai2)
Noometry Index30.532.6
Released2025-03-112024-11-21
WeightsOpenOpen
Context window131K—
Max output8K—
Input $ / M tokens$0.08—
Output $ / M tokens$0.16—
Results tracked4314

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

Category by category

Coding Tulu 3 (Tülu 3) 70B leads

Gemma 3 27B: 22.1 (#329), Tulu 3 (Tülu 3) 70B: 35.4 (#211)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Coding13221236
Aider Polyglot4.9%—
SciCode21.2%—
LiveBench Coding39.9%—

Agentic & Tool Use Not comparable

Gemma 3 27B: 24.9 (#105), Tulu 3 (Tülu 3) 70B: —

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
Berkeley Function Calling Leaderboard29.5%—

Reasoning Tulu 3 (Tülu 3) 70B leads

Gemma 3 27B: 16.3 (#297), Tulu 3 (Tülu 3) 70B: 22.9 (#172)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Hard Prompts13401220
Kagi LLM Benchmark40.4%—
CritPt0%—
Chess Puzzles0%—
LiveBench Reasoning43.8%—
DTBench52.5%—
LiveBench Data Analysis51.5%—
LMCA12.3%—
Epoch Capabilities Index130.04—
LiveBench50%—

Math Gemma 3 27B leads

Gemma 3 27B: 24.4 (#261), Tulu 3 (Tülu 3) 70B: 13.3 (#297)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
OTIS Mock AIME 2024-202522.5%4.4%
LMArena Math13121242
MATH Level 574%42.7%
LiveBench Math55.4%—

Knowledge Too close to call

Gemma 3 27B: 25.4 (#256), Tulu 3 (Tülu 3) 70B: 25.1 (#257)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
GPQA Diamond47.7%46.3%
Confabulations40.3%—
Vectara Hallucination Rate7.4%—
LMArena Expert1303—

Multimodal Not comparable

Gemma 3 27B: 32.0 (#94), Tulu 3 (Tülu 3) 70B: —

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Vision1165—
GeoBench52%—

Multilingual Gemma 3 27B leads

Gemma 3 27B: 47.0 (#150), Tulu 3 (Tülu 3) 70B: 40.1 (#217)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Non-English13341236
LMArena Chinese13471249
LMArena Russian13491246
LMArena French1368—
LMArena German1362—
LMArena Japanese1287—
LMArena Korean1308—
LMArena Spanish1348—

Instruction Following Gemma 3 27B leads

Gemma 3 27B: 70.8 (#155), Tulu 3 (Tülu 3) 70B: 65.0 (#222)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Instruction Following13211233
LiveBench Instruction Following74.9%—

Long Context Tulu 3 (Tülu 3) 70B leads

Gemma 3 27B: 27.6 (#288), Tulu 3 (Tülu 3) 70B: 37.1 (#217)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Longer Query13331224
Fiction.LiveBench33.3%—

Writing & Preference Gemma 3 27B leads

Gemma 3 27B: 52.7 (#161), Tulu 3 (Tülu 3) 70B: 45.8 (#218)

BenchmarkGemma 3 27BTulu 3 (Tülu 3) 70B
LMArena Text13581256
LMArena Creative Writing13461231
LMArena Multi-Turn13441252
Short-Story Creative Writing79.9%—
EQ-Bench Creative Writing1266—
LiveBench Language34.6%—

Frequently asked questions

Is Gemma 3 27B better than Tulu 3 (Tülu 3) 70B?

Tulu 3 (Tülu 3) 70B is the stronger model overall, scoring 32.6 to 30.5 on the Noometry Index.

Is Gemma 3 27B or Tulu 3 (Tülu 3) 70B better for coding?

Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 35.4 versus 22.1 in the Noometry coding category.

How many benchmarks do Gemma 3 27B and Tulu 3 (Tülu 3) 70B share?

14 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.

Related comparisons

Go deeper