Noometry

Model comparison

Grok 4.6 vs Llama 3-70B

Grok 4.6 is the stronger model overall, scoring 56.8 to 28.5 on the Noometry Index.

Last verified . 21 shared benchmarks.

Grok 4.6 xAI

56.8

Rank #19 Confirmed

Llama 3-70B Meta

28.5

Rank #317 Confirmed

Summary

  • They share 21 benchmarks with published results for both. Grok 4.6 scores higher in 9 categories and Llama 3-70B in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Grok 4.6 leads 66.8 to 12.0.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 99.2% for Grok 4.6 and 4.3% for Llama 3-70B.
  • Llama 3-70B has downloadable open weights; the other is API-only.

Side by side

Grok 4.6Llama 3-70B
ProviderxAIMeta
Noometry Index56.828.5
Released2026-08-122024-04-18
WeightsProprietaryOpen
Context window500K—
Max output500K—
Input $ / M tokens$2—
Output $ / M tokens$6—
Results tracked4931

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

Category by category

Coding Grok 4.6 leads

Grok 4.6: 58.0 (#14), Llama 3-70B: 35.3 (#215)

BenchmarkGrok 4.6Llama 3-70B
LMArena Coding14641206
DeepSWE67.5%—
FrontierCode48%—
CursorBench41.4%—
LMArena WebDev1620—
FrontierSWE25.3%—
SciCode56.5%—
WeirdML67.3%—
BigCodeBench Instruct—43.6%
BigCodeBench Complete—54.5%
ALE-Bench1,508—
HumanEval+—72%
MBPP+—69%

Agentic & Tool Use Grok 4.6 leads

Grok 4.6: 39.4 (#25), Llama 3-70B: 20.9 (#132)

BenchmarkGrok 4.6Llama 3-70B
APEX-Agents65.3%—
Cybench—5%
GDP.pdf17.2%—
Vending-Bench 29,047—

Reasoning Grok 4.6 leads

Grok 4.6: 61.3 (#18), Llama 3-70B: 17.4 (#284)

BenchmarkGrok 4.6Llama 3-70B
LMArena Hard Prompts14451195
DTBench97.3%54.2%
Epoch Capabilities Index156.48122.91
ARC-AGI-267.1%—
SimpleBench75.9%—
Kagi LLM Benchmark—35.1%
NYT Connections (extended)80%—
ARC-AGI-187.5%—
CritPt19.7%—
Chess Puzzles40%—
EBR-Bench30.5%—
Mystery Game Puzzles34%—
LMCA48.5%—
ForecastBench—57.1
WinoGrande—83.5%

Math Grok 4.6 leads

Grok 4.6: 66.8 (#19), Llama 3-70B: 12.0 (#299)

BenchmarkGrok 4.6Llama 3-70B
OTIS Mock AIME 2024-202599.2%4.3%
LMArena Math14161218
FrontierMath (Tiers 1-3)66%—
FrontierMath Tier 431.7%—
ProofBench51%—
MATH Level 5—22.6%

Knowledge Grok 4.6 leads

Grok 4.6: 63.3 (#18), Llama 3-70B: 20.8 (#272)

BenchmarkGrok 4.6Llama 3-70B
GPQA Diamond94%40.6%
LMArena Expert14631149
SimpleQA Verified49.3%—
MMLU—79.3%

Multimodal Not comparable

Grok 4.6: 43.9 (#18), Llama 3-70B: —

BenchmarkGrok 4.6Llama 3-70B
LMArena Vision1265—
Blueprint-Bench 233.2%—
Furniture Assembly40%—
LMArena Document1452—

Multilingual Grok 4.6 leads

Grok 4.6: 53.3 (#67), Llama 3-70B: 33.8 (#246)

BenchmarkGrok 4.6Llama 3-70B
LMArena Non-English14221142
LMArena Chinese14791114
LMArena French14581231
LMArena German14251169
LMArena Japanese13831017
LMArena Korean13951017
LMArena Russian14271159
LMArena Spanish14081241

Instruction Following Grok 4.6 leads

Grok 4.6: 75.6 (#59), Llama 3-70B: 62.8 (#233)

BenchmarkGrok 4.6Llama 3-70B
LMArena Instruction Following14311194

Long Context Grok 4.6 leads

Grok 4.6: 44.4 (#62), Llama 3-70B: 35.6 (#235)

BenchmarkGrok 4.6Llama 3-70B
LMArena Longer Query14531174

Writing & Preference Grok 4.6 leads

Grok 4.6: 62.5 (#77), Llama 3-70B: 43.0 (#225)

BenchmarkGrok 4.6Llama 3-70B
LMArena Text14271221
LMArena Creative Writing14301210
LMArena Multi-Turn14221223

Frequently asked questions

Is Grok 4.6 better than Llama 3-70B?

Grok 4.6 is the stronger model overall, scoring 56.8 to 28.5 on the Noometry Index.

Is Grok 4.6 or Llama 3-70B better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.0 versus 35.3 in the Noometry coding category.

How many benchmarks do Grok 4.6 and Llama 3-70B share?

21 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Llama 3-70B has 31.

Related comparisons

Go deeper