Instructions to use AIMS2025/DeepSignal with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AIMS2025/DeepSignal with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="AIMS2025/DeepSignal",
	filename="DeepSignal-Phase-4B_V1.F16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use AIMS2025/DeepSignal with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf AIMS2025/DeepSignal:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf AIMS2025/DeepSignal:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf AIMS2025/DeepSignal:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf AIMS2025/DeepSignal:Q4_K_M

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 AIMS2025/DeepSignal:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf AIMS2025/DeepSignal:Q4_K_M

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 AIMS2025/DeepSignal:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AIMS2025/DeepSignal:Q4_K_M

Use Docker

docker model run hf.co/AIMS2025/DeepSignal:Q4_K_M

LM Studio
Jan

vLLM

How to use AIMS2025/DeepSignal with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AIMS2025/DeepSignal"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AIMS2025/DeepSignal",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AIMS2025/DeepSignal:Q4_K_M

Ollama
How to use AIMS2025/DeepSignal with Ollama:
```
ollama run hf.co/AIMS2025/DeepSignal:Q4_K_M
```

Unsloth Studio new

How to use AIMS2025/DeepSignal with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AIMS2025/DeepSignal to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AIMS2025/DeepSignal to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for AIMS2025/DeepSignal to start chatting

Docker Model Runner
How to use AIMS2025/DeepSignal with Docker Model Runner:
```
docker model run hf.co/AIMS2025/DeepSignal:Q4_K_M
```

Lemonade

How to use AIMS2025/DeepSignal with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AIMS2025/DeepSignal:Q4_K_M

Run and chat with the model

lemonade run user.DeepSignal-Q4_K_M

List all available models

lemonade list

AIMS2025 commited on Apr 17

Commit

e32dbc1

verified ·

1 Parent(s): 60b62af

Rename CyclePlan GGUF files to V1

Browse files

Files changed (5) hide show

.gitattributes +3 -0
DeepSignal_CyclePlan-4B-F16.gguf → DeepSignal_CyclePlan-4B-V1-F16.gguf +0 -0
DeepSignal_CyclePlan-4B-Q4_K_M.gguf → DeepSignal_CyclePlan-4B-V1-Q4_K_M.gguf +0 -0
DeepSignal_CyclePlan-4B-Q8_0.gguf → DeepSignal_CyclePlan-4B-V1-Q8_0.gguf +0 -0
README.md +10 -10

.gitattributes CHANGED Viewed

@@ -41,3 +41,6 @@ DeepSignal-CyclePlan-4B_V1.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
 DeepSignal_CyclePlan-4B-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
 DeepSignal_CyclePlan-4B-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
 DeepSignal_CyclePlan-4B-F16.gguf filter=lfs diff=lfs merge=lfs -text

 DeepSignal_CyclePlan-4B-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
 DeepSignal_CyclePlan-4B-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
 DeepSignal_CyclePlan-4B-F16.gguf filter=lfs diff=lfs merge=lfs -text
+DeepSignal_CyclePlan-4B-V1-F16.gguf filter=lfs diff=lfs merge=lfs -text
+DeepSignal_CyclePlan-4B-V1-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+DeepSignal_CyclePlan-4B-V1-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text

DeepSignal_CyclePlan-4B-F16.gguf → DeepSignal_CyclePlan-4B-V1-F16.gguf RENAMED Viewed

File without changes

DeepSignal_CyclePlan-4B-Q4_K_M.gguf → DeepSignal_CyclePlan-4B-V1-Q4_K_M.gguf RENAMED Viewed

File without changes

DeepSignal_CyclePlan-4B-Q8_0.gguf → DeepSignal_CyclePlan-4B-V1-Q8_0.gguf RENAMED Viewed

File without changes

README.md CHANGED Viewed

@@ -21,20 +21,20 @@ This repository provides GGUF checkpoints for local inference with traffic-signa
 This repository currently contains:
 - **DeepSignal-Phase-4B-V1**: next signal-phase prediction
-- **DeepSignal_CyclePlan-4B**: cycle-level green-time allocation for all phases in the upcoming signal cycle
 ## Model Files
 | Filename | Task | Quantization | Size | Notes |
 |:---|:---|:---:|:---:|:---|
 | `DeepSignal-Phase-4B_V1.F16.gguf` | Phase prediction | F16 | ~8 GB | Existing phase model |
-| `DeepSignal_CyclePlan-4B-F16.gguf` | Cycle planning | F16 | ~7.5 GB | Highest-fidelity CyclePlan checkpoint |
-| `DeepSignal_CyclePlan-4B-Q8_0.gguf` | Cycle planning | Q8_0 | ~4.0 GB | Balanced quality / speed |
-| `DeepSignal_CyclePlan-4B-Q4_K_M.gguf` | Cycle planning | Q4_K_M | ~2.4 GB | Recommended for local inference |
-## DeepSignal_CyclePlan-4B
-`DeepSignal_CyclePlan-4B` is a traffic signal **cycle planning** model. It takes the predicted traffic state for the next cycle and outputs the green-time allocation for each phase while respecting phase-specific minimum and maximum green constraints.
 ### Recommended Prompt Format
@@ -62,7 +62,7 @@ Wrap the input with `【cycle_predict_input_json】...【/cycle_predict_input_js
 `Q4_K_M` is the recommended local default:
 ```bash
-llama-cli -m DeepSignal_CyclePlan-4B-Q4_K_M.gguf \
   --ctx-size 8192 \
   --temp 0.2 \
   -p 'You are a traffic signal timing optimization expert.
@@ -100,7 +100,7 @@ The final answer should contain a machine-readable plan inside `<SOLUTION>...</S
 ### Download Example
 ```bash
-huggingface-cli download AIMS2025/DeepSignal DeepSignal_CyclePlan-4B-Q4_K_M.gguf --local-dir .
 ```
 ## DeepSignal-Phase-4B-V1
@@ -135,9 +135,9 @@ where the number is the phase index (starting from 0) and the seconds is the dur
 | Model | Format Success Rate (%) | Avg Queue Vehicles | Avg Delay per Vehicle (s) | Throughput (veh/min) | Avg Response Time (s) |
 |:---:|:---:|:---:|:---:|:---:|:---:|
-| **DeepSignal_CyclePlan-4B F16 (thinking, Ours)** | **100.0** | **3.504** | **27.747** | **8.611** | 4.351 |
 | [`GLM-4.7-Flash (thinking)`](https://huggingface.co/zai-org/glm-4.7-flash) | 100.0 | 7.323 | 29.422 | 8.567 | 36.388 |
-| DeepSignal_CyclePlan-4B Q4_K_M (thinking, Ours) | 98.1 | 4.783 | 29.891 | 7.722 | 1.674 |
 | [`Qwen3-30B-A3B`](https://huggingface.co/Qwen/Qwen3-30B-A3B-2507) | 97.1 | 6.938 | 31.135 | 7.578 | 7.885 |
 | [`LightGPT-8B-Llama3`](https://huggingface.co/lightgpt/LightGPT-8B-Llama3) | 68.0 | 5.026 | 31.266 | 7.380 | 167.373 |
 | [`GPT-OSS-20B (thinking)`](https://huggingface.co/openai/gpt-oss-20b) | 65.4 | 6.289 | 31.947 | 7.247 | 4.919 |

 This repository currently contains:
 - **DeepSignal-Phase-4B-V1**: next signal-phase prediction
+- **DeepSignal_CyclePlan-4B-V1**: cycle-level green-time allocation for all phases in the upcoming signal cycle
 ## Model Files
 | Filename | Task | Quantization | Size | Notes |
 |:---|:---|:---:|:---:|:---|
 | `DeepSignal-Phase-4B_V1.F16.gguf` | Phase prediction | F16 | ~8 GB | Existing phase model |
+| `DeepSignal_CyclePlan-4B-V1-F16.gguf` | Cycle planning | F16 | ~7.5 GB | Highest-fidelity CyclePlan checkpoint |
+| `DeepSignal_CyclePlan-4B-V1-Q8_0.gguf` | Cycle planning | Q8_0 | ~4.0 GB | Balanced quality / speed |
+| `DeepSignal_CyclePlan-4B-V1-Q4_K_M.gguf` | Cycle planning | Q4_K_M | ~2.4 GB | Recommended for local inference |
+## DeepSignal_CyclePlan-4B-V1
+`DeepSignal_CyclePlan-4B-V1` is a traffic signal **cycle planning** model. It takes the predicted traffic state for the next cycle and outputs the green-time allocation for each phase while respecting phase-specific minimum and maximum green constraints.
 ### Recommended Prompt Format
 `Q4_K_M` is the recommended local default:
 ```bash
+llama-cli -m DeepSignal_CyclePlan-4B-V1-Q4_K_M.gguf \
   --ctx-size 8192 \
   --temp 0.2 \
   -p 'You are a traffic signal timing optimization expert.
 ### Download Example
 ```bash
+huggingface-cli download AIMS2025/DeepSignal DeepSignal_CyclePlan-4B-V1-Q4_K_M.gguf --local-dir .
 ```
 ## DeepSignal-Phase-4B-V1
 | Model | Format Success Rate (%) | Avg Queue Vehicles | Avg Delay per Vehicle (s) | Throughput (veh/min) | Avg Response Time (s) |
 |:---:|:---:|:---:|:---:|:---:|:---:|
+| **DeepSignal_CyclePlan-4B-V1 F16 (thinking, Ours)** | **100.0** | **3.504** | **27.747** | **8.611** | 4.351 |
 | [`GLM-4.7-Flash (thinking)`](https://huggingface.co/zai-org/glm-4.7-flash) | 100.0 | 7.323 | 29.422 | 8.567 | 36.388 |
+| DeepSignal_CyclePlan-4B-V1 Q4_K_M (thinking, Ours) | 98.1 | 4.783 | 29.891 | 7.722 | 1.674 |
 | [`Qwen3-30B-A3B`](https://huggingface.co/Qwen/Qwen3-30B-A3B-2507) | 97.1 | 6.938 | 31.135 | 7.578 | 7.885 |
 | [`LightGPT-8B-Llama3`](https://huggingface.co/lightgpt/LightGPT-8B-Llama3) | 68.0 | 5.026 | 31.266 | 7.380 | 167.373 |
 | [`GPT-OSS-20B (thinking)`](https://huggingface.co/openai/gpt-oss-20b) | 65.4 | 6.289 | 31.947 | 7.247 | 4.919 |