DiffuSpace Clover-35B-A3B-Preview: open diffusion-language model built for fast coding
A 35B-total, 3B-active mixture-of-experts coding model that generates text in blocks via parallel denoising instead of one token at a time, claiming 515 tokens/sec on an NVIDIA H200, released under Apache 2.0 on October 29.
Why it matters
It swaps autoregression for block diffusion with a Gated DeltaNet backbone and learnable block-size calibration, aiming at high-speed code generation from an open model rather than a hosted API.
What you could build with it
An indie developer could test this model as the draft engine inside a code assistant: generate whole function blocks in parallel, then run the project's own test suite as the gate, and compare wall-clock completion time against an autoregressive model of similar size before committing to the architecture.
Does it hold up?
Too early to judge: released October 29 as a preview with vendor-reported benchmarks only; the GGUF and MLX community conversions show immediate local-inference interest, but no independent coding evaluation exists yet.
Built with DiffuSpace Clover-35B-A3B-Preview
- DiffuSpace releases open-source Clover-35B modelarticle · Release-day report covering the block-diffusion design, 515 tok/s H200 claim, 65.3 on tau-cubed and 67.9 on BFCL-v4, and Apache 2.0 licensing.
- maternion/Clover-35B-A3B-Preview-GGUFgithub · Community GGUF quant of the preview weights for local inference runtimes.
- kapitsaprime/Clover-35B-A3B-Preview-MLX-4bitgithub · Community 4-bit MLX conversion aimed at Apple Silicon inference.
Learn more
First spotted on hf: source.
More AI models
Tencent Hunyuan Hy3: open 295B MoE with 21B active under Apache 2.0Tencent open-sourced Hy3, a 295B-parameter mixture-of-experts model with only 21B parameters active per token, under…models · JEV 0.73NVIDIA Isaac Lab 3.0 GA + Isaac OS open runtime + Agile ONE SN humanoidMajor Isaac platform expansion at GTC Berlin Oct 21: Isaac Lab 3.0 is generally available as the open-source robot…models · JEV 0.73Kimi K3.1 Pro: Moonshot's 2.8T open-weight flagship officially launchesMoonshot AI officially launched Kimi K3.1 Pro on Oct 23, a mixture-of-experts model with 2.8 trillion total / 104…models · JEV 0.72NVIDIA Nemotron 3.5 Retriever collection: Embed 8B/1B, Rerank 8B, Parse 631MNVIDIA's Oct 21 collection drop for agentic retrieval: Nemotron-3.5-Embed in 8B BF16, 1B BF16 and 1B NVFP4…models · JEV 0.72NVIDIA Cosmos 3.5 world models: now openly availableNVIDIA's Cosmos 3.5 family of open world foundation models for robots, autonomous vehicles and vision agents is now…models · JEV 0.71NVIDIA RFD4-Proteina: open protein-design BioNeMo modelOpen protein-design model from NVIDIA and the Baker Lab at UW Medicine's Institute for Protein Design, announced at…models · JEV 0.7
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.