MDL
DeepSeek Coder V2 Lite Instruct
DeepSeek
Verified against Artificial Analysis · Jul 13, 2026
Capability
44
percentile index
Price /M
$0 / $0
in / out
Context
—
Avg score
22.0
5 benchmarks
AA Intelligence
3.1
index
Weights
Open
Released
Jun 17, 2024
About
DeepSeek Coder V2 Lite Instruct is a open-weights model in the DeepSeek family from DeepSeek. Benchmarked on 5 evals, averaging 22.0.
Benchmark scores · 5
avg 22.0 — higher is betterThe research behind this model
all papers mentioning it →FPMoE: A Sparse Mixture-of-Experts Approach to Functional Code GenerationMay 27, 2026Delulu: A Verified Multi-Lingual Benchmark for Code Hallucination Detection in Fill-in-the-Middle TasksMay 7, 2026Delta-Based Neural Architecture Search: LLM Fine-Tuning via Code DiffsMay 6, 2026From Code to Prediction: Fine-Tuning LLMs for Neural Network Performance Classification in NNGPTMay 5, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$0/M