Instructions to use TensorFold/Qwen3.8-Flash-Next-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TensorFold/Qwen3.8-Flash-Next-MLX-4bit with MLX:

# Make sure mlx-vlm is installed
# pip install --upgrade mlx-vlm

from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template
from mlx_vlm.utils import load_config

# Load the model
model, processor = load("TensorFold/Qwen3.8-Flash-Next-MLX-4bit")
config = load_config("TensorFold/Qwen3.8-Flash-Next-MLX-4bit")

# Prepare input
image = ["http://images.cocodataset.org/val2017/000000039769.jpg"]
prompt = "Describe this image."

# Apply chat template
formatted_prompt = apply_chat_template(
    processor, config, prompt, num_images=1
)

# Generate output
output = generate(model, processor, formatted_prompt, image)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use TensorFold/Qwen3.8-Flash-Next-MLX-4bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "TensorFold/Qwen3.8-Flash-Next-MLX-4bit"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "TensorFold/Qwen3.8-Flash-Next-MLX-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use TensorFold/Qwen3.8-Flash-Next-MLX-4bit with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "TensorFold/Qwen3.8-Flash-Next-MLX-4bit"

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 TensorFold/Qwen3.8-Flash-Next-MLX-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use TensorFold/Qwen3.8-Flash-Next-MLX-4bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "TensorFold/Qwen3.8-Flash-Next-MLX-4bit"

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 "TensorFold/Qwen3.8-Flash-Next-MLX-4bit" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

ashxhart commited on Aug 26

Commit

de59776

verified ·

1 Parent(s): 932f42f

Add M3 Studio performance and MTP compatibility note

Browse files

Files changed (1) hide show

README.md +16 -1

README.md CHANGED Viewed

@@ -58,6 +58,9 @@ The upstream tokenizer, chat template, vision processor, and generation configur
 > [!IMPORTANT]
 > Qwen3.8 Flash Next uses the new `qwen4_exp` architecture. Use an oMLX or MLX-VLM build that explicitly lists `qwen4_exp` support. Older MLX-VLM releases cannot load this checkpoint.
 ## Quick start
 ```bash
@@ -74,6 +77,18 @@ python -m mlx_vlm.generate \
   --max-tokens 512
 ```
 ## Architecture
 Qwen3.8 Flash Next is an experimental vision-language architecture combining Gated DeltaNet, Qwen Sparse Attention, sparse mixture-of-experts layers, widened gated residual streams, and hashed bigram/trigram embeddings.
@@ -95,7 +110,7 @@ For upstream evaluations, intended use, limitations, safety guidance, and the co
 - Source: official BF16 checkpoint.
 - All 3,671 converted tensors and 22 indexed shards were checked locally.
 - The release payload was scanned for credentials, personal contact details, private paths, private network information, logs, caches, and private organisation data.
-- Generation performance will be added after the repeatable Apple-silicon benchmark finishes.
 - Quantisation can reduce output quality relative to BF16. Test the model on representative workloads before production use.
 This is a community conversion, not an official Qwen release.

 > [!IMPORTANT]
 > Qwen3.8 Flash Next uses the new `qwen4_exp` architecture. Use an oMLX or MLX-VLM build that explicitly lists `qwen4_exp` support. Older MLX-VLM releases cannot load this checkpoint.
+> [!CAUTION]
+> Do not attach a Qwen3.8 27B MTP drafter to this model. The hidden sizes differ and the drafter is incompatible with Flash Next.
 ## Quick start
 ```bash
   --max-tokens 512
 ```
+## Measured performance
+Validated on an Apple M3 Studio with text-only generation after model load:
+| Test path | Result |
+| --- | ---: |
+| oMLX server, warmed 543–566-token responses | 24.1–24.2 tokens/s |
+| oMLX server, warmed shorter responses | 24.6–26.1 tokens/s |
+| Standalone MLX exact-copy smoke test | 31.0 tokens/s |
+The standalone result is a short smoke test; the longer oMLX figures better represent sustained chat generation. Results vary with prompt length, cache state, sampling settings, runtime version, and memory pressure.
 ## Architecture
 Qwen3.8 Flash Next is an experimental vision-language architecture combining Gated DeltaNet, Qwen Sparse Attention, sparse mixture-of-experts layers, widened gated residual streams, and hashed bigram/trigram embeddings.
 - Source: official BF16 checkpoint.
 - All 3,671 converted tensors and 22 indexed shards were checked locally.
 - The release payload was scanned for credentials, personal contact details, private paths, private network information, logs, caches, and private organisation data.
+- Deterministic standalone and warmed oMLX server generation tests passed on Apple silicon.
 - Quantisation can reduce output quality relative to BF16. Test the model on representative workloads before production use.
 This is a community conversion, not an official Qwen release.