Clover-35B-A3B-Preview — CMF

DiffuSpace Clover-35B-A3B-Preview — a diffusion language model on the Qwen3.6-35B-A3B backbone (mixture of experts: 35B parameters, ~3B active per token, 256 experts top-8 plus a shared expert; 30 Gated DeltaNet and 10 gated full-attention layers; 128k context). It is built to write in 32-token blocks by repeated denoising passes, and its trunk also reads as an ordinary causal model — packaged as single CMF files for cortiq, a Rust inference engine with no Python and no ML framework (CUDA, Vulkan, Metal and the CPU). cortiq decodes causally by default, the fastest on every machine we measured; block denoising and self-speculation are one setting away.

English · Русский · 中文

Quick start

cargo install cortiq-cli
hf download infosave/Clover-35B-A3B-Preview-cmf clover-35b-a3b-preview-q4tp-v2.cmf --local-dir .
cortiq run clover-35b-a3b-preview-q4tp-v2.cmf --prompt "Write a Python function that merges two sorted lists." --no-think --max-tokens 512

Use cortiq 0.8.52 or later; prebuilt binaries for Linux, macOS and Windows are on the releases page. Without --no-think the model reasons in a <think> block first, so give it --max-tokens in the thousands.

Files

file experts size wiki ppl first-pass top-1 recommended for
clover-35b-a3b-preview-q4tp-v2.cmf 4-bit, GPTQ 19.36 GB 16.80 93/96 default: GPUs with 24 GB+
clover-35b-a3b-preview-q4tp.cmf 4-bit 19.38 GB 18.22 93/96 the plain conversion
clover-35b-a3b-preview-q8_2f.cmf 8-bit 35.00 GB 15.86 62/64 near-lossless; GPUs with 48 GB+, big-memory CPUs

Attention, Gated DeltaNet projections, embeddings and the head are 8-bit in every file. q4tp-v2 calibrates the experts with GPTQ on 262k tokens (wiki, chat and code text) at the same size as q4tp. wiki ppl: perplexity of the model's causal path over 12 × 512 tokens of wikitext-2 test. first-pass top-1: rows of the first denoising pass whose top token matches the official PyTorch implementation (bf16), three chat prompts × 32 rows (two for q8_2f); the misses are near-ties.

Performance

Tokens per second after the prompt, q4tp-v2:

hardware prose code a 2048-token prompt
RTX 5090 (32 GB), Vulkan 120 121 1143 tok/s
16 GB of VRAM, Vulkan + CPU 54 53
Mac mini M4 (24 GB), prompt on Metal, decode on CPU 13–18 14–18 ~170 tok/s
CPU only, 15 Xeon cores 13.8 10.9

The whole model and its head take 18.7 GB on the card. q8_2f does not fit 32 GB whole and runs at 46–49 tok/s there.

RTX A5000 (24 GB), CUDA, cortiq 0.8.96, cortiq bench (CMF_GPU=cuda on a build with the CUDA backend (cargo install cortiq-cli --features cuda, the NVIDIA driver and the CUDA 12.x libraries)): q4tp-v2 decodes at 111.5–112.5 tok/s and reads a 4096-token prompt at 2356 tok/s (first token 1.7 s); Vulkan on the same card gives 76 tok/s and 445 tok/s (9.2 s).

Decoding modes

CMF_CLOVER_MODE picks how a request uses the model. On the RTX 5090:

mode what it does prose code
ar (default) causal decode with the release's response penalties 120 121
draft self-speculation: a denoising pass proposes 16 tokens, the causal path verifies them 42 104
denoise LowConfidence block denoising, the release's generate 18–26 63–95

In denoise on an M4: 3.2 and 6.5–9 tok/s. CMF_CLOVER_THRESHOLD lowers the denoising threshold, CMF_CLOVER_DRAFT sets the rows of a self-speculative round.

Usage

cortiq run clover-35b-a3b-preview-q4tp-v2.cmf                                  # interactive chat
cortiq run clover-35b-a3b-preview-q4tp-v2.cmf --prompt "..." --max-tokens 4096 # with reasoning
cortiq run clover-35b-a3b-preview-q4tp-v2.cmf --prompt "..." --no-think        # direct answer

Every mode applies the release's response penalties (presence 0.9, repetition 1.05, a gradual push toward the stop token after 8192 answer tokens and toward </think> after 4096). --temperature samples instead of greedy.

OpenAI-compatible server

cortiq serve clover-35b-a3b-preview-q4tp-v2.cmf --port 8080

curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "clover", "enable_thinking": false,
       "messages": [{"role": "user", "content": "Explain how a hash map works."}]}'

/v1/chat/completions (streamed or not), /v1/completions, /v1/models and a web dashboard on the same port.

Hardware

NVIDIA / AMD / Intel (Vulkan, DX12). A card that holds the file runs the whole model; a smaller one runs the first layers on the card and the rest on the CPU. cortiq gpu lists the adapters; on headless Linux install libvulkan1 libglvnd0 libegl1 libgl1 libglx0 and set XDG_RUNTIME_DIR=/tmp.

Apple silicon. The prompt runs on Metal and the decode on the CPU, which is the faster of the two for this file on a 24 GB Mac: Metal cannot keep the 19 GB of weights resident there. The weights stay in the file's mapping, so a 24 GB Mac runs the q4tp files.

CPU only. CMF_GPU=0.

Verify the download

sha256sum -c clover-35b-a3b-preview-q4tp-v2.cmf.sha256
cortiq verify clover-35b-a3b-preview-q4tp-v2.cmf
cortiq info clover-35b-a3b-preview-q4tp-v2.cmf

Reproduce

cortiq convert --model Diffuspace/Clover-35B-A3B-Preview --quant q4tp \
  --sampling temperature=0,top_p=1,top_k=0,min_p=0,repetition_penalty=1 \
  --output clover-35b-a3b-preview-q4tp.cmf
cortiq quantize-moe clover-35b-a3b-preview-q4tp.cmf --hf Diffuspace/Clover-35B-A3B-Preview \
  --calib wiki.train.txt=0.4 --calib chat.txt=0.4 --calib code.txt=0.2 \
  --tokens 262144 --codec q4tp --output clover-35b-a3b-preview-q4tp-v2.cmf

Scope

Causal decoding, LowConfidence block diffusion and self-speculation. Weights derive from DiffuSpace's release (built on Qwen3.6-35B-A3B) and remain under its Apache-2.0 terms; see the upstream card and its attributions.


Документация на русском

Clover-35B-A3B-Preview от DiffuSpace — диффузионная языковая модель на основе Qwen3.6-35B-A3B (35 млрд параметров, ~3 млрд активных на токен). Она умеет писать блоками по 32 токена, раскрывая их проходами денойзинга, а её основа работает и как обычная каузальная модель. Здесь она в виде отдельных файлов CMF для движка cortiq на Rust; по умолчанию cortiq декодирует каузально: это быстрее всего на всех машинах, где мы мерили.

cargo install cortiq-cli
hf download infosave/Clover-35B-A3B-Preview-cmf clover-35b-a3b-preview-q4tp-v2.cmf --local-dir .
cortiq run clover-35b-a3b-preview-q4tp-v2.cmf --prompt "Объясни, как работает хеш-таблица." --no-think --max-tokens 512

Нужен cortiq 0.8.52 или новее. Файл по умолчанию — q4tp-v2 (эксперты 4 бита с GPTQ, 19.36 ГБ, wiki ppl 16.80); q4tp — простая конвертация (18.22), q8_2f — почти без потерь (35 ГБ, 15.86).

Токены в секунду после промпта, q4tp-v2: RTX 5090 (Vulkan) — 120 на прозе и 121 на коде, промпт на 2048 токенов читается со скоростью 1143 ток/с; при 16 ГБ видеопамяти — 54 и 53; Mac mini M4 (24 ГБ, промпт на Metal ~170 ток/с, декод на CPU) — 13–18 и 14–18; только CPU, 15 ядер Xeon — 13.8 и 10.9.

RTX A5000 (24 ГБ) с CUDA (cortiq 0.8.96, cortiq bench, сборка --features cuda, CMF_GPU=cuda): q4tp-v2 генерирует 111.5–112.5 ток/с и читает промпт в 4096 токенов со скоростью 2356 ток/с (первый токен через 1.7 с).

CMF_CLOVER_MODE выбирает режим: ar (по умолчанию) — каузальный декод со штрафами из релиза; draft — самоспекуляция, проход денойзинга предлагает 16 токенов, каузальный путь их проверяет (RTX 5090: 42 и 104 ток/с); denoise — блочный денойзинг LowConfidence из релиза (RTX 5090: 18–26 и 63–95; M4: 3.2 и 6.5–9). cortiq serve даёт OpenAI-совместимый API с потоковой выдачей.

中文说明

Clover-35B-A3B-Preview 是 DiffuSpace 基于 Qwen3.6-35B-A3B 的扩散语言模型(35B 参数,每个 token 约 3B 激活)。它可以用 32 个 token 的块经多次去噪生成,其主干也可作 普通因果模型运行。本仓库提供可由 Rust 推理引擎 cortiq(0.8.52 及以上)加载的 CMF 文件;默认文件为 q4tp-v2(专家 4 位 GPTQ,19.36 GB)。cortiq 默认因果解码,这在 我们测过的所有机器上都最快。

每秒 token 数(提示之后,q4tp-v2):RTX 5090(Vulkan)文本 120、代码 121,2048 个 token 的提示 1143 token/s;16 GB 显存 54 和 53;Mac mini M4(24 GB,提示走 Metal 约 170 token/s,解码走 CPU)13–18 和 14–18;仅 CPU(15 核 Xeon)13.8 和 10.9。

RTX A5000(24 GB)上使用 CUDA(cortiq 0.8.96,cortiq bench,--features cuda 构建,CMF_GPU=cuda):q4tp-v2 解码 111.5–112.5 token/s,4096 token 的提示 2356 token/s(首 token 1.7 秒)。

CMF_CLOVER_MODE 选择模式:ar(默认)为因果解码;draft 为自推测,一次去噪提出 16 个 token,由因果路径验证(RTX 5090:42 和 104);denoise 为发布版的 LowConfidence 块去噪(RTX 5090:18–26 和 63–95;M4:3.2 和 6.5–9)。

hf download infosave/Clover-35B-A3B-Preview-cmf clover-35b-a3b-preview-q4tp-v2.cmf --local-dir .
cortiq run clover-35b-a3b-preview-q4tp-v2.cmf --prompt "你好" --no-think
Downloads last month
439
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for infosave/Clover-35B-A3B-Preview-cmf

Quantized
(22)
this model