Instructions to use MediaStreamAI/MOTHER_CORE_V2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MediaStreamAI/MOTHER_CORE_V2 with Transformers:

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

pipe = pipeline("text-generation", model="MediaStreamAI/MOTHER_CORE_V2", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("MediaStreamAI/MOTHER_CORE_V2", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MediaStreamAI/MOTHER_CORE_V2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MediaStreamAI/MOTHER_CORE_V2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MediaStreamAI/MOTHER_CORE_V2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MediaStreamAI/MOTHER_CORE_V2

SGLang

How to use MediaStreamAI/MOTHER_CORE_V2 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 "MediaStreamAI/MOTHER_CORE_V2" \
    --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": "MediaStreamAI/MOTHER_CORE_V2",
		"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 "MediaStreamAI/MOTHER_CORE_V2" \
        --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": "MediaStreamAI/MOTHER_CORE_V2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MediaStreamAI/MOTHER_CORE_V2 with Docker Model Runner:
```
docker model run hf.co/MediaStreamAI/MOTHER_CORE_V2
```

MediaStreamAI commited on 20 days ago

Commit

809df71

verified ·

1 Parent(s): 5112f48

Correct training hyperparameters: SEQ=4096 (not 2048), GRAD_ACCUM_STEPS=32 (not 8). Training is at full architecture context length; no RoPE extrapolation needed for inference.

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -298,8 +298,8 @@ Forward return:
 |---|---|
 | Learning rate | 1e-5 |
 | Gradient clip | 10.0 |
-| Effective batch size | 8 (BATCH_PHYSICAL=1 × GRAD_ACCUM_STEPS=8) |
-| Sequence length (training) | 2048 |
 | Optimiser | AdamW (β₁=0.9, β₂=0.95) |
 | Weight decay | 0.1 |
 | Warmup steps | 100 |
@@ -307,7 +307,7 @@ Forward return:
 | Hardware | NVIDIA GB10 Blackwell (Grace–Blackwell unified memory, 128GB) |
 | Training site | MSAI Wright Avenue, Dundee — sovereign UK infrastructure |
-Training was performed at sequence length **2048** using physical microbatches of 1 with gradient accumulation of 8 (effective batch = 8). The architecture supports 4,096-token inference; 2048 → 4096 is a modest RoPE extrapolation, but long-context behaviour at full 4096 has not been benchmarked at this checkpoint.
 ---

 |---|---|
 | Learning rate | 1e-5 |
 | Gradient clip | 10.0 |
+| Effective batch size | 32 (BATCH_PHYSICAL=1 × GRAD_ACCUM_STEPS=32) |
+| Sequence length (training) | 4096 |
 | Optimiser | AdamW (β₁=0.9, β₂=0.95) |
 | Weight decay | 0.1 |
 | Warmup steps | 100 |
 | Hardware | NVIDIA GB10 Blackwell (Grace–Blackwell unified memory, 128GB) |
 | Training site | MSAI Wright Avenue, Dundee — sovereign UK infrastructure |
+Training was performed at the full architecture sequence length of **4096** using physical microbatches of 1 with gradient accumulation of 32 (effective batch = 32). Because training and inference share the same context length, no RoPE extrapolation is required for 4096-token inference. Long-context behaviour at full 4096 has been exposed during training but not formally benchmarked at this checkpoint.
 ---