Text Generation
Transformers
Safetensors
English
cma
custom_code
causal-lm
small-language-model
generalist
4k-tokenizer
Instructions to use SupraLabs/SupraCMA-8M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use SupraLabs/SupraCMA-8M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="SupraLabs/SupraCMA-8M", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("SupraLabs/SupraCMA-8M", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use SupraLabs/SupraCMA-8M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "SupraLabs/SupraCMA-8M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "SupraLabs/SupraCMA-8M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/SupraLabs/SupraCMA-8M
- SGLang
How to use SupraLabs/SupraCMA-8M 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 "SupraLabs/SupraCMA-8M" \ --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": "SupraLabs/SupraCMA-8M", "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 "SupraLabs/SupraCMA-8M" \ --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": "SupraLabs/SupraCMA-8M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use SupraLabs/SupraCMA-8M with Docker Model Runner:
docker model run hf.co/SupraLabs/SupraCMA-8M
Add inverse theme support to visuals
Browse files- README.md +2 -2
- assets/cma-architecture.svg +39 -12
- assets/training-trajectory.svg +26 -6
README.md
CHANGED
|
@@ -54,7 +54,7 @@ CMA in the feed-forward position. For every token, CMA reshapes the 288-channel
|
|
| 54 |
hidden state into **12 chunks of 24**, routes between those chunks with **3
|
| 55 |
heads**, applies a signed blend and SiLU gate, then projects back to 288 channels.
|
| 56 |
|
| 57 |
-

|
|
| 137 |
decreased from **3.7838** to **1.3520**, while the Open SLM Leaderboard-style
|
| 138 |
average increased from **30.66%** to **36.21%**.
|
| 139 |
|
| 140 |
-

|
| 58 |
|
| 59 |
## Quick start
|
| 60 |
|
|
|
|
| 137 |
decreased from **3.7838** to **1.3520**, while the Open SLM Leaderboard-style
|
| 138 |
average increased from **30.66%** to **36.21%**.
|
| 139 |
|
| 140 |
+

|
| 141 |
|
| 142 |
## Evaluation
|
| 143 |
|
assets/cma-architecture.svg
CHANGED
|
|
|
|
assets/training-trajectory.svg
CHANGED
|
|
|
|