AION
Modelofficial release

deepseek-ai/DeepSeek-V3-Base

DeepSeek models →Hugging Face ↗
Parameters685B
Context160K
Input → output–
License–
PrecisionF8_E4M3 · fp8
ArchitectureDeepseekV3ForCausalLM
ReleasedDec 25, 2024
Downloads4.3K

Other sizes and formats in the family

ModelParamsPrecisionDownloads
deepseek-ai/DeepSeek-V3fp8685BF8_E4M31.5M

Compare side by side →

This release in the news

No coverage of this exact release yet.

Recent news about DeepSeek models

  1. Oct 11, 2026 · Opinion / analysis
    PSA: DeepSeek V4.1 Flash habitually exfiltrates API keys. It is dangerously misaligned and may be hazardous to use
  2. Oct 8, 2026 · Research paper
    A Closer Look at Agentic BBO: Benchmarking LLM Agents for Black-Box Optimization
  3. Oct 8, 2026 · Research paper
    Internalizer: Portable Context-to-Parameter Mapping for Very Large Language Models
  4. Oct 8, 2026 · Research paper
    Mine Odyssey: Benchmarking Spatial Agentic Intelligence in the Wild
  5. Oct 7, 2026 · Research paper
    StoreBench: A Live-Commerce Environment for Evaluating and Training Autonomous Operator Agents
  6. Oct 7, 2026 · Research paper
    Evaluating Rubric Generation with Interventional Transfer
  7. Oct 6, 2026 · Research paper
    GeoNatureAgent (GNA): A Framework and Benchmark for Pre-Production Evaluation of Tool-Using Agents on Geospatial and Environmental Tasks
  8. Oct 6, 2026 · Research paper
    Beyond the Leaderboard: Multi-Dimensional Evaluation of Dense and Mixture-of-Experts Models for Automated Program Repair