Noometry

Model comparison

Gemini 3.1 Pro Preview vs Llama 3.1-70B

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 29.4 on the Noometry Index. Llama 3.1-70B costs 11× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Last verified . 25 shared benchmarks.

Gemini 3.1 Pro Preview Google

56.7

Rank #20 Confirmed

Llama 3.1-70B Meta

29.4

Rank #299 Confirmed

Summary

  • They share 25 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 9 categories and Llama 3.1-70B in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 20.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 95.6% for Gemini 3.1 Pro Preview and 3.6% for Llama 3.1-70B.
  • Llama 3.1-70B is cheaper at $0.40 / $0.40 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 128K.
  • Llama 3.1-70B has downloadable open weights; the other is API-only.

Side by side

Gemini 3.1 Pro PreviewLlama 3.1-70B
ProviderGoogleMeta
Noometry Index56.729.4
Released2026-02-192024-07-23
WeightsProprietaryOpen
Context window1.05M128K
Max output66K4K
Input $ / M tokens$2$0.40
Output $ / M tokens$12$0.40
Results tracked7135

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

Category by category

Coding Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 42.7 (#88), Llama 3.1-70B: 29.8 (#292)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
WeirdML72.1%9%
LMArena Coding14831260
SWE-bench Verified75.6%—
DeepSWE11.7%—
LMArena WebDev1446—
SciCode58.9%—
GSO22.6%—
BigCodeBench Instruct—46.1%
MirrorCode8.9%—
BigCodeBench Complete—54.8%
ALE-Bench1,161—
AlgoTune2.02—

Agentic & Tool Use Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 37.7 (#33), Llama 3.1-70B: 24.9 (#104)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
BALROG57%27.9%
Terminal-Bench80.2%—
APEX-Agents35.3%—
TheAgentCompany—6.9%
τ²-bench Banking26%—
DeepResearch Bench47.8%—
PostTrainBench22%—
ExploitBench26.1%—
GBAEval0.8%—
GDP.pdf17%—
LMArena Search1211—
METR Time Horizons77%—
Vending-Bench 23,774—

Reasoning Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.7 (#9), Llama 3.1-70B: 20.8 (#215)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
LMArena Hard Prompts14841241
DTBench97.1%60%
LMCA53.8%14.8%
Epoch Capabilities Index154.92125.91
ARC-AGI-277.1%—
SimpleBench79.6%—
NYT Connections (extended)97.4%—
ARC-AGI-198%—
CritPt17.7%—
Chess Puzzles55%—
EnigmaEval36.8%—
Thematic Generalization79.4%—
EBR-Bench14.3%—
Mystery Game Puzzles34%—
ForecastBench59—

Math Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 61.8 (#28), Llama 3.1-70B: 12.7 (#298)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
OTIS Mock AIME 2024-202595.6%3.6%
LMArena Math14851252
FrontierMath (Tiers 1-3)59.6%—
FrontierMath Tier 426.8%—
MathArena Final-Answer Competitions86.5%—
ProofBench26%—
Omni-MATH—21%
MATH Level 5—36.7%
FrontierMath (Feb 2025 set)36.9%—
FrontierMath Tier 4 (v1)16.7%—

Knowledge Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.8 (#3), Llama 3.1-70B: 24.1 (#264)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
GPQA Diamond94.4%44.2%
LMArena Expert14841209
Humanity's Last Exam46.4%—
SimpleQA Verified73.5%—
MMLU-Pro—65.3%
Vectara Hallucination Rate10.4%—
GPQA (HELM)—42.6%
MMLU—80.1%

Multimodal Not comparable

Gemini 3.1 Pro Preview: 38.1 (#60), Llama 3.1-70B: —

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
LMArena Vision1296—
Blueprint-Bench 226.5%—
Furniture Assembly26.7%—
LMArena Document1444—

Multilingual Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 57.1 (#11), Llama 3.1-70B: 38.9 (#220)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
LMArena Non-English14751219
LMArena Chinese15281215
LMArena French14881260
LMArena German14911222
LMArena Japanese14941132
LMArena Korean14581140
LMArena Russian14961234
LMArena Spanish14801253

Instruction Following Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 77.1 (#31), Llama 3.1-70B: 65.5 (#218)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
LMArena Instruction Following14651231
IFEval—82.1%

Long Context Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 47.3 (#17), Llama 3.1-70B: 37.6 (#208)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
LMArena Longer Query14821241
CL-bench20.8%—
CL-bench Life16.9%—

Writing & Preference Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 66.3 (#36), Llama 3.1-70B: 35.6 (#262)

BenchmarkGemini 3.1 Pro PreviewLlama 3.1-70B
LMArena Text14801261
LMArena Creative Writing14821232
EQ-Bench Creative Writing1491784
LMArena Multi-Turn14861256
WildBench—75.8%
EQ-Bench 41142—

Frequently asked questions

Is Gemini 3.1 Pro Preview better than Llama 3.1-70B?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 29.4 on the Noometry Index. Llama 3.1-70B costs 11× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Which is cheaper, Gemini 3.1 Pro Preview or Llama 3.1-70B?

Llama 3.1-70B is cheaper. It lists at $0.40 per million input tokens and $0.40 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

Is Gemini 3.1 Pro Preview or Llama 3.1-70B better for coding?

Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.7 versus 29.8 in the Noometry coding category.

Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 128K.

How many benchmarks do Gemini 3.1 Pro Preview and Llama 3.1-70B share?

25 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Llama 3.1-70B has 35.

Related comparisons

Go deeper