MDL
DeepSeek-V3
deepseek-ai
Capability
79
percentile index
Downloads
1.1M
on HF
Likes
4.1K
Avg score
82.6
6 benchmarks
Weights
Open
Released
Dec 25, 2024
About
DeepSeek-V3 is a open-weights model in the DeepSeek family from deepseek-ai. Benchmarked on 6 evals, averaging 82.6.
text-generation
Benchmark scores · 6
avg 82.6 — higher is betterThe research behind this model
all papers mentioning it →DeepSeek-V3 Technical ReportEEVEE: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents▲ 18 on HF · Jun 9, 2026OptSkills: Learning Generalizable Optimization Skills from Problem Archetypes via Cluster-Based DistillationMay 28, 2026How Far Can Disaggregation Go? A Design-Space Exploration of Attention-FFN Disaggregation for Efficient MoE LLM ServingMay 27, 2026Better Accuracies, Worse Reasoning: A Step-Level Audit of Medical Chain-of-Thought DistillationMay 27, 2026TCP-MCP: Landscape-Guided Co-Evolution of Prompts and Communication Topologies for Multi-Agent SystemsMay 27, 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.