Noometry

Model comparison

Grok 4.20 (Non-Reasoning) vs Llama-3.3-70B-Instruct

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 10× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

Last verified . 24 shared benchmarks.

Grok 4.20 (Non-Reasoning) xAI

48.6

Rank #54 Confirmed

Llama-3.3-70B-Instruct Meta

30.6

Rank #291 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 9 categories and Llama-3.3-70B-Instruct in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 14.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Grok 4.20 (Non-Reasoning) and 5.1% for Llama-3.3-70B-Instruct.
  • Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
  • Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 128K.
  • Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.

Side by side

Grok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
ProviderxAIMeta
Noometry Index48.630.6
Released2026-02-172024-12-06
WeightsProprietaryOpen
Context window1M128K
Max output30K4K
Input $ / M tokens$1.25$0.10
Output $ / M tokens$2.50$0.32
Results tracked4643

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

Category by category

Coding Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 42.1 (#112), Llama-3.3-70B-Instruct: 31.0 (#290)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
WeirdML52.3%14.4%
LMArena Coding14591268
LMArena WebDev1375—
SciCode—26%
BigCodeBench Instruct—46.9%
LiveBench Coding—36.6%
BigCodeBench Complete—57.5%
ALE-Bench1,150—

Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 34.4 (#46), Llama-3.3-70B-Instruct: 25.8 (#105)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
Terminal-Bench57.3%—
Berkeley Function Calling Leaderboard—31.9%
τ²-bench Banking18%—
BALROG—23%
LMArena Search1189—
Vending-Bench 24,663—

Reasoning Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.3 (#32), Llama-3.3-70B-Instruct: 14.1 (#327)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
LMArena Hard Prompts14511257
DTBench90.1%59.5%
LMCA38.7%17.5%
Epoch Capabilities Index151.98127.33
ForecastBench61.458.6
ARC-AGI-265.1%—
SimpleBench—19.9%
Kagi LLM Benchmark75%—
NYT Connections (extended)85.4%—
ARC-AGI-189.5%—
CritPt—0%
Chess Puzzles24%—
Thematic Generalization63.8%—
LiveBench Reasoning—50.8%
LiveBench Data Analysis—49.5%
LiveBench—50.2%

Math Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 48.2 (#65), Llama-3.3-70B-Instruct: 15.3 (#298)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
OTIS Mock AIME 2024-202592.2%5.1%
LMArena Math14551267
FrontierMath (Tiers 1-3)44.9%—
FrontierMath Tier 417.1%—
ProofBench14%—
LiveBench Math—42.2%
MATH Level 5—41.6%

Knowledge Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.8 (#60), Llama-3.3-70B-Instruct: 30.6 (#226)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
GPQA Diamond89.3%47.4%
LMArena Expert14391225
SimpleQA Verified30.2%—
Confabulations—22.8%
Vectara Hallucination Rate—4.1%
MMLU—86.3%

Multimodal Not comparable

Grok 4.20 (Non-Reasoning): 33.3 (#98), Llama-3.3-70B-Instruct: —

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
LMArena Vision1263—
Blueprint-Bench 20%—
LMArena Document1416—

Multilingual Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 54.5 (#40), Llama-3.3-70B-Instruct: 39.9 (#220)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
LMArena Non-English14411236
LMArena Chinese14811217
LMArena French14761281
LMArena German14651251
LMArena Japanese14491150
LMArena Korean14171143
LMArena Russian14581252
LMArena Spanish14431270

Instruction Following Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 74.8 (#83), Llama-3.3-70B-Instruct: 71.1 (#157)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
LMArena Instruction Following14201242
LiveBench Instruction Following—82.7%

Long Context Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 45.5 (#34), Llama-3.3-70B-Instruct: 26.4 (#295)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
LMArena Longer Query14371256
Fiction.LiveBench—33.3%
CL-bench22.2%—
CL-bench Life11.9%—

Writing & Preference Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 65.7 (#44), Llama-3.3-70B-Instruct: 47.6 (#207)

BenchmarkGrok 4.20 (Non-Reasoning)Llama-3.3-70B-Instruct
LMArena Text14511274
LMArena Creative Writing14381250
LMArena Multi-Turn14561280
EQ-Bench Creative Writing1574—
LiveBench Language—39.2%

Frequently asked questions

Is Grok 4.20 (Non-Reasoning) better than Llama-3.3-70B-Instruct?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 10× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

Which is cheaper, Grok 4.20 (Non-Reasoning) or Llama-3.3-70B-Instruct?

Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

Is Grok 4.20 (Non-Reasoning) or Llama-3.3-70B-Instruct better for coding?

Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 42.1 versus 31.0 in the Noometry coding category.

Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 128K.

How many benchmarks do Grok 4.20 (Non-Reasoning) and Llama-3.3-70B-Instruct share?

24 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Llama-3.3-70B-Instruct has 43.

Related comparisons

Go deeper