Instructions to use tg-techie-agents/Q1D-4B-Blk0-V1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use tg-techie-agents/Q1D-4B-Blk0-V1 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 tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0
# Run inference directly in the terminal:
llama cli -hf tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0
# Run inference directly in the terminal:
llama cli -hf tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

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 tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0
# Run inference directly in the terminal:
./llama-cli -hf tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

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 tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

Use Docker

docker model run hf.co/tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

LM Studio
Jan

vLLM

How to use tg-techie-agents/Q1D-4B-Blk0-V1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tg-techie-agents/Q1D-4B-Blk0-V1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tg-techie-agents/Q1D-4B-Blk0-V1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

Ollama
How to use tg-techie-agents/Q1D-4B-Blk0-V1 with Ollama:
```
ollama run hf.co/tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0
```
Unsloth Desktop
Docker Model Runner
How to use tg-techie-agents/Q1D-4B-Blk0-V1 with Docker Model Runner:
```
docker model run hf.co/tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0
```

Lemonade

How to use tg-techie-agents/Q1D-4B-Blk0-V1 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tg-techie-agents/Q1D-4B-Blk0-V1:Q1_0

Run and chat with the model

lemonade run user.Q1D-4B-Blk0-V1-Q1_0

List all available models

lemonade list

Atomic Chat

Q1D 4B · Block 0 · V1

A research artifact from Q1 Descent — reconstructing 1-bit "intelligence-density" recovery for open-weight models.

What it is. Qwen3-4B with transformer block 0 quantized to 1-bit (Q1_0_g128: 1.125 bpw, a sign per weight + one FP16 scale per 128-group) and then trained back toward the full-precision teacher. Blocks 1–35 remain F16. So it isolates a single 1-bit block in an otherwise-F16 model — a validation of the recovery recipe, not a fully 1-bit model.

Recipe (V1). Straight-through estimator: FP32 master weights, forward = the exact Q1_0 encoding (so both signs and scales move), backward = straight through. Objective: KL to the FP16 teacher on C4 — 400 steps, lr 1e-4, bf16. The trained weights are the deployed 1-bit encoding (no latent/deploy gap).

Result (GSM8K, greedy, thinking-off):

block 0 = 1-bit, rest F16	GSM8K
naive (untrained Q1_0)	0.25
trained (this model)	0.93
F16 reference	0.94

Training recovers the dented block to ≈ full-precision capability at 1 bit.

Use. llama.cpp / LM Studio. Ships a closed-think (no-think) chat template; greedy (temperature 0) recommended.

hf download tg-techie-agents/Q1D-4B-Blk0-V1

Caveats. Early research artifact: one 1-bit block (not a full 1-bit model), 4B scratchpad scale, single seed, GSM8K-only. Known token-level instability in some contexts (an occasional malformed token) — under investigation. Not affiliated with PrismML or the Qwen team.

Downloads last month: 15

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

1-bit

Model tree for tg-techie-agents/Q1D-4B-Blk0-V1

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Quantized

(315)

this model