Hugging Face 热门模型第 30 名：Diffuspace/Clover-35B-A3B-Preview，144 赞，近期下载 1030 次。 标签：transformers, safetensors, clover_moe, text-generation, clover, mixture-of-experts, diffusion, custom-code, conversational, custom_code, base_model:Qwen/Qwen3.6-35B-A3B, base_model:finetune:Qwen/Qwen3.6-35B-A3B

README：

Clover-35B-A3B-Preview

Clover-35B-A3B-Preview is the first model released by DiffuSpace, delivering state-of-the-art performance among open-source diffusion language models at release on the benchmarks reported in our launch blog. Clover combines a 35B-parameter mixture-of-experts backbone with approximately 3B active parameters per token, block diffusion generation, and a 128K context window.

Blog · Code & inference backends · Hugging Face

Performance

See the benchmark discussion for evaluation results, settings, and comparisons.

Architecture and decoding

Clover is a text-only model derived from Qwen3.6-35B-A3B, retaining its hybrid Gated DeltaNet (GDN) / full-attention MoE backbone with 40 layers, 256 routed experts, top-8 token routing, and one shared expert per layer.

• Bidirectional block attention: additional learned GDN parameters, including right-convolution weights and softmax controls, let tokens attend within a noisy block while preserving a causal prefix.
• 128K context: up to 131,072 tokens for the prompt and generated response together.
• Block diffusion with LowConfidence decoding: this Hugging Face implementation starts each block with masked tokens and resolves confident positions in parallel. Positions with confidence above the threshold are filled; if none qualify, the most confident position is filled to ensure progress. Once resolved, the block becomes part of the prefix for the next block. Defaults are block_size=32 and threshold=0.95.

For self-speculative decoding with our SGLang and llama.cpp implementations, see the official code repository.

Quick start

Install PyTorch with CUDA support, then the Hugging Face dependencies. The example below uses Transformers 5.19.0 and PyTorch 2.10 or later.

pip install "transformers==5.19.0" accelerate safetensors

Use a GPU with enough memory for the BF16 weights (approximately 69 GB) plus inference state.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Diffuspace/Clover-35B-A3B-Preview"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    trust_remote_code=True,
    dtype=torch.bfloat16,
    device_map="cuda:0",
).eval()

messages = [{"role": "user", "content": "What is diffusion?"}]
prompt = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True,
)
inputs = tokenizer(prompt, add_special_tokens=False, return_tensors="pt").to("cuda:0")
output = model.generate(
    **inputs,
    max_new_tokens=24576,
    block_size=32,
    threshold=0.95,
    do_sample=False,
)
completion = output[0, inputs["input_ids"].shape[1]:]
print(tokenizer.decode(completion, skip_special_tokens=True))

You can also set model_id to a local model directory. Generation defaults are loaded automatically from generation_config.json; keyword arguments to generate() override them. Requests exceeding the 128K context limit are rejected, including space needed to resolve the final full block.

Generation controls

We observe that very long generations can sometimes become repetitive. The model includes repetition controls and gradual biases toward ending the response or thinking phase. The defaults are already configured, so no extra arguments are needed for normal use. They also apply during greedy decoding.

Parameter	Default	Purpose
presence_penalty	0.9	Discourages tokens that have already appeared in the generated response.
frequency_penalty	0.0	Penalizes tokens in proportion to their occurrence count; disabled by default.
repetition_penalty	1.05	Reduces the preference for previously generated tokens; 1.0 disables it.
dllm_length_penalty	10.0	Gradually increases the preference for end-of-response tokens.
dllm_think_end_bias	10.0	Gradually encourages `` to close the thinking phase.

Length control begins at dllm_length_penalty_start=8192 response tokens and reaches its full strength over dllm_length_penalty_window=8192 more tokens. Thinking control uses dllm_think_end_start=4096 and dllm_think_end_window=4096. These are soft controls; max_new_tokens sets the hard output limit. Repetition controls operate on generated response history, excluding the prompt.

Expert selection

During block denoising, Clover can restrict routing to a shared subset of experts for each block. This reduces the number of distinct experts used across the block. The relevant model configuration settings are:

Setting	Default	Purpose
moe_block_coreset_enabled	true	Enables block-level expert selection during denoising.
moe_block_coreset_top_p	0.9	Sets the cumulative routing-mass threshold used to select the expert subset.

The selector aggregates the routing scores of each token's top-8 candidate experts across the block, ranks experts by their aggregate score, and

The original belongs to Hugging Face 热门模型 and its authors and is reproduced here for reading. To have it removed, write to support@aisa.one.