Noometry

Model comparison

Deepseek Coder v2 vs DeepSeek-R1-Distill-Llama-70B

DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.4 to 35.3 on the Noometry Index.

Last verified . 2 shared benchmarks.

Deepseek Coder v2 DeepSeek

35.3

Rank #217 Confirmed

DeepSeek-R1-Distill-Llama-70B DeepSeek

37.4

Rank #187 Confirmed

Summary

  • They share 2 benchmarks with published results for both. Deepseek Coder v2 scores higher in 2 categories and DeepSeek-R1-Distill-Llama-70B in 4 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where DeepSeek-R1-Distill-Llama-70B leads 49.0 to 38.4.
  • The biggest single-benchmark swing is BigCodeBench Instruct: 48.2% for Deepseek Coder v2 and 35.3% for DeepSeek-R1-Distill-Llama-70B.

Side by side

Deepseek Coder v2DeepSeek-R1-Distill-Llama-70B
ProviderDeepSeekDeepSeek
Noometry Index35.337.4
Released2024-06-172025-01-20
WeightsOpenOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked2413

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

Category by category

Coding Deepseek Coder v2 leads

Deepseek Coder v2: 37.5 (#178), DeepSeek-R1-Distill-Llama-70B: 36.3 (#197)

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
BigCodeBench Instruct48.2%35.3%
BigCodeBench Complete59.7%49.9%
LiveBench Coding—51.6%
LMArena Coding1251—
HumanEval+82.3%—
MBPP+75.1%—

Reasoning DeepSeek-R1-Distill-Llama-70B leads

Deepseek Coder v2: 22.6 (#175), DeepSeek-R1-Distill-Llama-70B: 24.1 (#150)

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
Kagi LLM Benchmark—52.3%
LiveBench Reasoning—67.6%
LMArena Hard Prompts1207—
LiveBench Data Analysis—55.9%
LiveBench—54.5%
WinoGrande83.7%—

Math DeepSeek-R1-Distill-Llama-70B leads

Deepseek Coder v2: 32.3 (#188), DeepSeek-R1-Distill-Llama-70B: 34.3 (#161)

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
OTIS Mock AIME 2024-2025—51.4%
LiveBench Math—58.1%
LMArena Math1241—
MATH Level 5—89.9%
GSM8K94.5%—

Knowledge Deepseek Coder v2 leads

Deepseek Coder v2: 31.7 (#209), DeepSeek-R1-Distill-Llama-70B: 30.7 (#216)

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
GPQA Diamond—55.7%
LMArena Expert1181—
ARC (AI2) Challenge64.3%—

Multilingual Not comparable

Deepseek Coder v2: 36.4 (#235), DeepSeek-R1-Distill-Llama-70B: —

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
LMArena Non-English1182—
LMArena Chinese1201—
LMArena French1184—
LMArena German1164—
LMArena Japanese1126—
LMArena Korean1104—
LMArena Russian1188—
LMArena Spanish1153—

Instruction Following DeepSeek-R1-Distill-Llama-70B leads

Deepseek Coder v2: 61.9 (#237), DeepSeek-R1-Distill-Llama-70B: 68.3 (#185)

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
LiveBench Instruction Following—69.9%
LMArena Instruction Following1180—

Long Context Not comparable

Deepseek Coder v2: 37.0 (#219), DeepSeek-R1-Distill-Llama-70B: —

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
LMArena Longer Query1219—

Writing & Preference DeepSeek-R1-Distill-Llama-70B leads

Deepseek Coder v2: 38.4 (#249), DeepSeek-R1-Distill-Llama-70B: 49.0 (#189)

BenchmarkDeepseek Coder v2DeepSeek-R1-Distill-Llama-70B
LMArena Text1191—
LMArena Creative Writing1120—
LMArena Multi-Turn1176—
LiveBench Language—23.8%

Frequently asked questions

Is Deepseek Coder v2 better than DeepSeek-R1-Distill-Llama-70B?

DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.4 to 35.3 on the Noometry Index.

Is Deepseek Coder v2 or DeepSeek-R1-Distill-Llama-70B better for coding?

Deepseek Coder v2 scores higher on coding benchmarks: 37.5 versus 36.3 in the Noometry coding category.

How many benchmarks do Deepseek Coder v2 and DeepSeek-R1-Distill-Llama-70B share?

2 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and DeepSeek-R1-Distill-Llama-70B has 13.

Related comparisons

Go deeper