Instructions to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF
# Run inference directly in the terminal:
llama cli -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF
# Run inference directly in the terminal:
llama cli -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF
# Run inference directly in the terminal:
./llama-cli -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Use Docker

docker model run hf.co/nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

LM Studio
Jan

vLLM

How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Ollama
How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with Ollama:
```
ollama run hf.co/nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF
```
Unsloth Desktop

How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with Docker Model Runner:
```
docker model run hf.co/nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF
```

Lemonade

How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Run and chat with the model

lemonade run user.Qwen3.6-35B-A3B-APEX-MTP-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Qwen3.6-35B-A3B APEX I-Balanced + Native MTP head(Lynn 推荐 server-grade 本地版)

这是 Qwen3.6-35B-A3B 的 APEX I-Balanced 量化 + MTP head 完整保留版,在 Lynn GB10 Spark 上实测 think-on 长链 75-85 TPS(无 MTP baseline ~66 TPS),适合 24GB+ 显存 / 32GB+ 内存的本地高质量推理。

定位:端侧 35B 主力,质量接近 BF16,MTP 加速 30% 在 thinking 场景。

English summary: this is the Qwen3.6-35B-A3B model with APEX adaptive-precision MoE quantization (I-Balanced variant) with the vendor-shipped MTP (Multi-Token Prediction) head preserved for use with llama.cpp --spec-type draft-mtp. Lynn-benchmarked on NVIDIA GB10 Spark (sm_121). Pairs with the lighter Lynn-9B-GGUF-imatrix release for the Lynn 4-quadrant matrix's heavyweight local tier.

文件 / Files

文件	大小	SHA256	备注
`Qwen3.6-35B-A3B-APEX-MTP-I-Balanced.gguf`	24.27 GiB(26.06 GB)	`9bf7d96bb3a9d363e645dd998aee9e9bff8e016a82aec7ff081e0e6cdb53419e`	APEX I-Balanced 量化 + MTP head(`blk.40.nextn.eh_proj/enorm/hnorm/shared_head_norm`)

量化来源:本仓库 GGUF 来自社区 APEX-quant 流程(adaptive-precision MoE),Lynn 未重新量化,只做 Spark TPS / 集成验证 + 双语 README + 发布镜像。原始量化产物归属社区 APEX-quant project。

为什么选这一档

35B-A3B(35B 总参数 / 3B active expert)是 Qwen 在 2025 端侧推理性价比之王:

active 3B params → decode 速度跟 dense 3-4B 接近(GB10 Spark 单流 60+ TPS)
total 35B + 256 experts → quality 大幅超越同 active 参数 dense 模型
APEX I-Balanced 量化:重要 expert 高精度,边缘 expert 更激进,体积 24GB 适合 32GB+ 内存设备
vendor MTP head 保留:llama.cpp --spec-type draft-mtp 自动启用 single-stream speculative decoding,thinking 长链路再 +29%

TPS 实测 / Throughput Benchmarks

NVIDIA GB10 Spark(sm_121), llama.cpp build-cuda-sm121 commit b1-5d44db6, 2026-05-23。Single-stream decode TPS via llama.cpp built-in predicted_per_second timing。

本仓库(APEX-MTP,`--spec-type draft-mtp` 启用)

Scenario	Cap	Tokens generated	Decode TPS	Notes
think-off 短答	1024	628	59.70	typical chat reply(MTP 在短答场景略损耗,见解读)
think-off 中长	2048	1808	61.95	sustained essay
think-on 数学题	4096	2145	84.69 ⭐	thinking reasoning
think-on 归纳证明	16384	4213	75.53 ⭐	math proof + Bayesian framing

vs 同硬件 APEX no-MTP 基线

Scenario	APEX no-MTP	APEX-MTP(本仓库)	Δ
think-off 1024	66.42	59.70	-10% ⚠️(MTP overhead > 收益)
think-on 4096	65.78	84.69	+29% ⭐
think-on 16K	~65 estimate	75.53	+16%

⚠️ 何时启用 MTP

启用:thinking_enabled=true 长链路 reasoning,coding 长输出,长 prompt → +20-30% TPS。

关闭 (--spec-type none):short chitchat / think-off 短答 / tool-call only → MTP 反而 -10%,因为 high-confidence n-gram 命中率低。

Lynn client v0.80+ 会根据 chat_template_kwargs.enable_thinking flag 自动 toggle。

评测摘要 / Benchmark Summary

Vendor official Qwen3.6-35B-A3B BF16:

MMLU 5-shot: 86.40
GPQA Diamond thinking-on 32K: ~86 (vendor reports)
AIME / Math: 80+

Lynn 内部 Q4_K_M-imatrix 同模型实测(memory reference_qwen36_35b_release_numbers_20260519):

MMLU 500 thinking-off: 83.00%
GPQA Diamond 198 thinking-off: 50.00%
thinking-on 32K 数据更高(待补)

APEX-MTP-I-Balanced quality 跟 Q4_K_M-imatrix 同档(都是 ~Q4 bit-depth),MTP head 只加速,不改 quality。

本地使用 / Local Usage

modelscope download --model Merkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF \
  Qwen3.6-35B-A3B-APEX-MTP-I-Balanced.gguf \
  --local_dir ~/Models/Lynn/Qwen3.6-35B-A3B

llama-server \
  --model ~/Models/Lynn/Qwen3.6-35B-A3B/Qwen3.6-35B-A3B-APEX-MTP-I-Balanced.gguf \
  --host 127.0.0.1 \
  --port 18098 \
  --ctx-size 32768 \
  --parallel 1 \
  --n-gpu-layers 999 \
  --cache-type-k q8_0 --cache-type-v q8_0 \
  --jinja --reasoning auto \
  --spec-type draft-mtp \
  --spec-draft-n-max 4 \
  --flash-attn on

显存预算:模型 24GB + KV cache(32K ctx, q8_0)~8GB ≈ 32GB GPU/unified mem。

OpenAI-compatible endpoint:

base_url = http://127.0.0.1:18098/v1
api_key  = local
model    = qwen36-35b-a3b-apex-mtp

MTP 验证 / MTP Verification

llama-server 启动 log 出现以下行确认 MTP active:

load_model: creating MTP draft context against the target model
common_speculative_impl_draft_mtp: adding speculative implementation 'draft-mtp'
  n_max=4, n_min=0, p_min=0.00, n_embd=2048, backend_sampling=1
speculative decoding context initialized

n_embd=2048 是 Qwen3.6-35B-A3B active expert hidden size(9B 同 head 是 n_embd=4096)。

来源 / Provenance

基座 Qwen3.6-35B-A3B:Alibaba Qwen 团队官方发布,Apache 2.0
APEX 量化:社区 mudler/apex-quant "Adaptive Precision for MoE" 流程,I-Balanced variant(重要 expert 高精度 / 边缘 expert 激进)
MTP head:Qwen 官方训练,保留进 GGUF
llama.cpp MTP 支持:upstream src/models/qwen35moe.cpp 的 graph_mtp 实现
Lynn 工作:Spark TPS benchmark + 集成 + 双语 README + MS/HF 镜像发布

License

Apache-2.0,inherits from base model Qwen/Qwen3.6-35B-A3B. APEX quantized weights distributed under the same Apache 2.0 license per base model terms.

Last updated: 2026-05-23. v1 APEX-MTP release with Lynn TPS benchmarks. Lynn project: https://github.com/MerkyorLynn/Lynn

Downloads last month: 76

GGUF

Model size

36B params

Architecture

qwen35moe

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

Base model

Qwen/Qwen3.6-35B-A3B

Quantized

(797)

this model

Qwen3.6-35B-A3B APEX I-Balanced + Native MTP head(Lynn 推荐 server-grade 本地版)

文件 / Files

为什么选这一档

TPS 实测 / Throughput Benchmarks

本仓库(APEX-MTP,--spec-type draft-mtp 启用)

vs 同硬件 APEX no-MTP 基线

⚠️ 何时启用 MTP

评测摘要 / Benchmark Summary

本地使用 / Local Usage

MTP 验证 / MTP Verification

来源 / Provenance

License

Model tree for nerkyor/Qwen3.6-35B-A3B-APEX-MTP-GGUF

本仓库(APEX-MTP,`--spec-type draft-mtp` 启用)