DeepSeek V3 Base
Newer version available: DeepSeek V4 Pro
Released
2024-12-26
Last refreshed
2026-10-17
Status
Researched today
Open sourceCommercial use: permittedLong context
DeepSeek V3 Base is a released long context model with open-source and 128K context; evaluate it while provider pricing coverage matures.
DeepSeek releases · 9 in the last 12 months · this family litChangelog →
Specifications
- Family
- DeepSeek V3
- Released
- 2024-12-26
- Context
- 128K
- Parameters
- 671B total, 37B active (MoE)
- Architecture
- Mixture of Experts
- Knowledge cutoff
- 2024-07
- Specialization
- general
- Openness
- Open source
- License
- MITOSI-approvedPermitted
- Weights
- Unknown
- Code
- Unknown
- Training
- Fine-tuned
Created by
Pricing
No tracked provider pricing is available yet.
About
DeepSeek V3 Base is DeepSeek's DeepSeek V3 model. It offers a 128K-token context window with weights openly available for self-hosting.
Providers
No active provider routes are tracked for this model yet.
Capabilities
No model capability flags are currently sourced.
Benchmark peer barsfor Long context
No task-mapped benchmark peers are available for this model yet.
Migration checks
No linked migration route is available for this model yet.
Compare DeepSeek V3 Base with other models
Comparison and alternatives
Browse all comparisons →DeepSeek V3 Base vs Hunyuan Hy3 PreviewDeepSeek V3 Base vs DeepSeek V4 FlashDeepSeek V3 Base vs Gemini 2.5 ProDeepSeek V3 Base vs GPT-5.2DeepSeek V3 Base vs GPT-5.2 CodexDeepSeek V3 Base vs o3DeepSeek V3 Base vs Mixtral 8x7BDeepSeek V3 Base vs GPT-4 TurboDeepSeek V3 Base vs Llama 3.1 405B InstructDeepSeek V3 Base vs Grok-3DeepSeek V3 Base vs GPT-5.5DeepSeek V3 Base vs Magistral Small 2506DeepSeek V3 Base vs Phi-4 Mini Flash ReasoningDeepSeek V3 Base vs Phi-4 Reasoning Vision 15B