Noometry

Model comparison

GPT-6 Astra vs Tulu 3 (Tülu 3) 70B

GPT-6 Astra is the stronger model overall, scoring 70.8 to 33.0 on the Noometry Index.

Last verified . 13 shared benchmarks.

GPT-6 Astra OpenAI

70.8

Rank #1 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. GPT-6 Astra 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 GPT-6 Astra leads 93.5 to 14.2.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for GPT-6 Astra 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

GPT-6 AstraTulu 3 (Tülu 3) 70B
ProviderOpenAIAllen Institute for AI (Ai2)
Noometry Index70.833.0
Released2026-09-032024-11-21
WeightsProprietaryOpen
Context window1.05M—
Max output128K—
Input $ / M tokens$10—
Output $ / M tokens$50—
Results tracked5614

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

Category by category

Coding GPT-6 Astra leads

GPT-6 Astra: 73.7 (#2), Tulu 3 (Tülu 3) 70B: 36.0 (#214)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Coding14871235
DeepSWE74.1%—
FrontierCode53.3%—
LMArena WebDev1786—
FrontierSWE65.5%—
SciCode56.5%—
GSO79.4%—
WeirdML93.6%—
MirrorCode46.7%—
ALE-Bench2,951—

Agentic & Tool Use Not comparable

GPT-6 Astra: 52.9 (#3), Tulu 3 (Tülu 3) 70B: —

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
APEX-Agents64.7%—
Remote Labor Index20.8%—
BALROG68.3%—
GDP.pdf34.2%—
Vending-Bench 215,515—

Reasoning GPT-6 Astra leads

GPT-6 Astra: 85.1 (#1), Tulu 3 (Tülu 3) 70B: 23.9 (#169)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Hard Prompts14621220
ARC-AGI-295%—
NYT Connections (extended)98.1%—
ARC-AGI-198.5%—
CritPt31.7%—
Chess Puzzles72%—
EBR-Bench76.2%—
Mystery Game Puzzles84%—
DTBench97.3%—
LMCA64.4%—
Bench to the Future 30.14—
Epoch Capabilities Index166.45—

Math GPT-6 Astra leads

GPT-6 Astra: 93.5 (#2), Tulu 3 (Tülu 3) 70B: 14.2 (#303)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
OTIS Mock AIME 2024-2025100%4.4%
LMArena Math14651242
FrontierMath (Tiers 1-3)93.7%—
FrontierMath Tier 497.6%—
ProofBench99%—
MATH Level 5—42.7%
FrontierMath Erdős2.9%—

Knowledge GPT-6 Astra leads

GPT-6 Astra: 75.3 (#1), Tulu 3 (Tülu 3) 70B: 25.0 (#264)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
GPQA Diamond95.8%46.3%
Humanity's Last Exam54.8%—
SimpleQA Verified75.6%—
Vectara Hallucination Rate8.7%—
LMArena Expert1483—

Multimodal Not comparable

GPT-6 Astra: 55.0 (#3), Tulu 3 (Tülu 3) 70B: —

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Vision1281—
Blueprint-Bench 249.7%—
Furniture Assembly80%—
LMArena Document1468—

Multilingual GPT-6 Astra leads

GPT-6 Astra: 53.7 (#61), Tulu 3 (Tülu 3) 70B: 39.9 (#222)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Non-English14301236
LMArena Chinese14841249
LMArena Russian14361246
LMArena French1456—
LMArena German1440—
LMArena Japanese1379—
LMArena Korean1426—
LMArena Spanish1407—

Instruction Following GPT-6 Astra leads

GPT-6 Astra: 76.3 (#44), Tulu 3 (Tülu 3) 70B: 64.8 (#227)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Instruction Following14501233

Long Context GPT-6 Astra leads

GPT-6 Astra: 44.5 (#62), Tulu 3 (Tülu 3) 70B: 37.1 (#222)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Longer Query14561224

Writing & Preference GPT-6 Astra leads

GPT-6 Astra: 75.3 (#7), Tulu 3 (Tülu 3) 70B: 45.6 (#223)

BenchmarkGPT-6 AstraTulu 3 (Tülu 3) 70B
LMArena Text14411256
LMArena Creative Writing14181231
LMArena Multi-Turn14481252
EQ-Bench Creative Writing2173—

Frequently asked questions

Is GPT-6 Astra better than Tulu 3 (Tülu 3) 70B?

GPT-6 Astra is the stronger model overall, scoring 70.8 to 33.0 on the Noometry Index.

Is GPT-6 Astra or Tulu 3 (Tülu 3) 70B better for coding?

GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 36.0 in the Noometry coding category.

How many benchmarks do GPT-6 Astra and Tulu 3 (Tülu 3) 70B share?

13 benchmarks have published results for both models. GPT-6 Astra has 56 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.

Related comparisons

Go deeper