Noometry

Model comparison

Llama-3.3-70B-Instruct vs StarCoder 2 7B

Llama-3.3-70B-Instruct has enough public results to be ranked (#291); StarCoder 2 7B does not yet, so treat this comparison as directional.

Last verified . 3 shared benchmarks.

Llama-3.3-70B-Instruct Meta

30.6

Rank #291 Confirmed

StarCoder 2 7B NVIDIA

37.0

Unranked Sparse

Summary

  • They share 3 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 0 categories and StarCoder 2 7B in 1 category; one gap is clear of the uncertainty.
  • The widest gap is in coding, where StarCoder 2 7B leads 34.8 to 31.0.
  • The biggest single-benchmark swing is BigCodeBench Complete: 57.5% for Llama-3.3-70B-Instruct and 27.7% for StarCoder 2 7B.

Side by side

Llama-3.3-70B-InstructStarCoder 2 7B
ProviderMetaNVIDIA
Noometry Index30.637.0
Released2024-12-062024-02-20
WeightsOpenOpen
Context window128K—
Max output4K—
Input $ / M tokens$0.10—
Output $ / M tokens$0.32—
Results tracked437

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

Category by category

Coding StarCoder 2 7B leads

Llama-3.3-70B-Instruct: 31.0 (#290), StarCoder 2 7B: 34.8

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
BigCodeBench Complete57.5%27.7%
SciCode26%—
WeirdML14.4%—
BigCodeBench Instruct46.9%—
LiveBench Coding36.6%—
LMArena Coding1268—
HumanEval+—29.9%

Agentic & Tool Use Not comparable

Llama-3.3-70B-Instruct: 25.8 (#105), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
Berkeley Function Calling Leaderboard31.9%—
BALROG23%—

Reasoning Not comparable

Llama-3.3-70B-Instruct: 14.1 (#327), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
Epoch Capabilities Index127.3393.56
SimpleBench19.9%—
CritPt0%—
LiveBench Reasoning50.8%—
LMArena Hard Prompts1257—
DTBench59.5%—
LiveBench Data Analysis49.5%—
LMCA17.5%—
ForecastBench58.6—
LiveBench50.2%—
WinoGrande—57.1%

Math Not comparable

Llama-3.3-70B-Instruct: 15.3 (#298), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
OTIS Mock AIME 2024-20255.1%—
LiveBench Math42.2%—
LMArena Math1267—
MATH Level 541.6%—
GSM8K—32.7%

Knowledge Not comparable

Llama-3.3-70B-Instruct: 30.6 (#226), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
MMLU86.3%38.8%
GPQA Diamond47.4%—
Confabulations22.8%—
Vectara Hallucination Rate4.1%—
LMArena Expert1225—
ARC (AI2) Challenge—38.7%

Multilingual Not comparable

Llama-3.3-70B-Instruct: 39.9 (#220), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
LMArena Non-English1236—
LMArena Chinese1217—
LMArena French1281—
LMArena German1251—
LMArena Japanese1150—
LMArena Korean1143—
LMArena Russian1252—
LMArena Spanish1270—

Instruction Following Not comparable

Llama-3.3-70B-Instruct: 71.1 (#157), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
LiveBench Instruction Following82.7%—
LMArena Instruction Following1242—

Long Context Not comparable

Llama-3.3-70B-Instruct: 26.4 (#295), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
Fiction.LiveBench33.3%—
LMArena Longer Query1256—

Writing & Preference Not comparable

Llama-3.3-70B-Instruct: 47.6 (#207), StarCoder 2 7B: —

BenchmarkLlama-3.3-70B-InstructStarCoder 2 7B
LMArena Text1274—
LMArena Creative Writing1250—
LMArena Multi-Turn1280—
LiveBench Language39.2%—

Frequently asked questions

Is Llama-3.3-70B-Instruct better than StarCoder 2 7B?

Llama-3.3-70B-Instruct has enough public results to be ranked (#291); StarCoder 2 7B does not yet, so treat this comparison as directional.

Is Llama-3.3-70B-Instruct or StarCoder 2 7B better for coding?

StarCoder 2 7B scores higher on coding benchmarks: 34.8 versus 31.0 in the Noometry coding category.

How many benchmarks do Llama-3.3-70B-Instruct and StarCoder 2 7B share?

3 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and StarCoder 2 7B has 7.

Related comparisons

Go deeper