DeepSeek AI · MoE Architecture
DeepSeek-V4-Pro-0813
A 1.7-trillion-parameter MoE with a DSpark speculative-decoding module and a 1M-token context, released MIT. Posts 87.9 on Terminal-Bench 2.1, 62.7 on DeepSWE, 74.1 on Toolathlon-Verified and 60.0 on HLE with tools. Exposes three reasoning-effort levels and will emit up to 384K output tokens in the higher modes.
Open the source ↗Similar Language Models releases
- Clover-35B-A3B-Preview
DiffuSpace · Diffusion MoE LLM · Week 45, 2026
- Groot-280B-A16B-261023
GrootTeam · Open reasoning MoE · Week 45, 2026
- Apex-35B-V1.0
Apex Intelligence · Research automation LLM · Week 45, 2026
- Ling-3.0-flash-Sante
inclusionAI (Ant Group) · Medical LLM · Week 44, 2026
- MemT-1.0-Nano
MemTensor · Memory LLM · Week 44, 2026
- Kimi K3.1 Pro
Moonshot AI · Open-Weight Frontier · Week 44, 2026
More from DeepSeek AI
- DeepSeek vision API
DeepSeek AI · Visual Understanding · Week 34, 2026
- DeepSeek Harness (dsh)
DeepSeek AI · Agent Framework · Week 33, 2026
- DeepSeek-V4-Flash-0731
DeepSeek AI · Reasoning · Week 31, 2026
- DeepSeek-OCR-2
DeepSeek AI · OCR & Document · Week 5, 2026
- DeepSeek-Math-V2
DeepSeek AI · Reasoning · Week 49, 2025
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.