Linkdeepseek-aiResource
The model brings competitive agentic and coding performance alongside out-of-the-box speculative decoding support in production engines like vLLM and SGLang. Released under the permissive MIT licence, it provides a strong foundation for cost-efficient self-hosted deployments.
With the 0731 release, DeepSeek focuses heavily on inference throughput and agentic execution. Incorporating speculative decoding directly into the primary weights via DSpark streamlines infrastructure setup by eliminating standalone draft models. Benchmark numbers show performance close to proprietary models such as Opus-4.8 on coding and terminal tasks. However, relying on custom encoding scripts instead of standard Jinja templates requires teams to adapt their prompt preprocessing pipelines.
Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.
Similar in content, found by the AI search.
X post:Mia
DeepSeek has officially published the weights for DeepSeek-V4-Flash-0731 under an MIT licence on Hugging Face. The release supersedes the preview version, features enhanced agentic capabilities, and incorporates an attached speculative decoding module.
336bookmarks153.4kviews
Link:deepseek-ai
DeepSeek has published the model card and weights for DeepSeek-V4-Pro-DSpark on Hugging Face. The release equips the DeepSeek-V4-Pro MoE model with an attached speculative decoding module (DSpark) for accelerated inference.
Article:DeepSeek AI
DeepSeek has open-sourced the preview release of DeepSeek-V4. The line-up features DeepSeek-V4-Pro and DeepSeek-V4-Flash, standardising a 1M context window and introducing architectural innovations to cut compute and memory overhead.
Link:deepseek-ai
DeepSeek-V4-Flash-0731 is a sparse Mixture-of-Experts language model featuring 13B active parameters and a 1-million-token context window. The NVIDIA Build model card outlines technical specifications, architecture details, and coding benchmark performances.
Article:Simon Willison
DeepSeek has introduced DeepSeek-V4-Pro and DeepSeek-V4-Flash under an MIT licence. Simon Willison reviews their Mixture of Experts architecture, KV cache efficiency gains, and pricing compared to OpenAI, Google, and Anthropic.
Link:Tencent Hy Team
The Tencent Hy Team has released Hy4-preview, an open-source Mixture-of-Experts (MoE) model featuring 770B total parameters and 49B activated parameters per token. The flagship model supports a 1M token context window and is tuned for software engineering, office analytics, and scientific research.
Parts of this content were created with the assistance of AI.
