Instructions to use dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8")
model = AutoModelForCausalLM.from_pretrained("dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8

SGLang

How to use dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8 with Docker Model Runner:
```
docker model run hf.co/dominant-strategies/quai-deepseek-v4-flash-igemm-fp4fp8
```

quai-deepseek-v4-flash-igemm-fp4fp8 / README.md

0xalank

Upload Quai Inference DeepSeek-V4-Flash IGEMM FP4/FP8 quantized checkpoint

d6a06bb verified 4 months ago

preview code

Raw History Blame Contribute Delete

2.15 kB

	---
	license: mit
	base_model: deepseek-ai/DeepSeek-V4-Flash
	library_name: transformers
	pipeline_tag: text-generation
	tags:
	- deepseek
	- deepseek-v4
	- fp4
	- fp8
	- moe
	- tensor-parallel
	- expert-parallel
	- inference-gemm
	- tensor-work-proof
	- quai
	---

	# Quai Inference DeepSeek-V4-Flash IGEMM FP4/FP8

	Built for Quai Inference.

	This repository contains InferenceGemm/Tensor Work Proof verifier-facing checkpoint artifacts derived from
	`deepseek-ai/DeepSeek-V4-Flash`. It is not an official DeepSeek release.

	Artifact precision: `FP4 experts + FP8 non-expert weights`.
	Artifact format: `quai-igemm-fp4fp8-v1`.

	## Contents

	- `igemm-*.safetensors`: Quai receipt-native tensor payloads.
	- `manifest.json`: verifier-facing tensor manifest and Merkle roots.
	- `index.json`: checkpoint index for the quantized tensor shards.
	- `runtime-alias-audit.json`, when present: runtime-to-manifest alias audit.
	- DeepSeek FP4/FP8 manifest audit and benchmark evidence, when present.
	- `benchmark-evidence/preflight/`, when present: preflight serving and
	receipt-path smoke evidence.
	- `docs/`, when present: Quai production receipt plan/status notes.
	- tokenizer/config files copied from the base model directory.

	## Benchmark Status

	Pending DeepSeek-V4-Flash production receipt benchmark. This repo is intended for the strict MoE+tensor-parallel Tensor Work Receipt path; do not promote until strict production gates and Go verification pass.

	The paper benchmark rows were produced with strict SGLang Tensor Work Receipt
	emission and Go verification. The raw benchmark logs, receipts, and local
	evidence packet are intentionally not uploaded to this model repository.

	## Loading

	These files are intended for the Quai InferenceGemm harness in this repository,
	not vanilla `transformers` weight loading:

	```text
	checkpoints/<this-checkpoint>/
	```

	Use the base model tokenizer/config with the InferenceGemm quantized payloads
	and verifier manifest.

	## Upstream

	- Base model: `deepseek-ai/DeepSeek-V4-Flash`
	- Upstream license: `mit`

	Redistribution must comply with the upstream model license and applicable export
	control restrictions.