Instructions to use ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64 with Transformers:

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

pipe = pipeline("text-generation", model="ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

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

tokenizer = AutoTokenizer.from_pretrained("ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64")
model = AutoModelForCausalLM.from_pretrained("ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64

SGLang

How to use ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64 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 "ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64 with Docker Model Runner:
```
docker model run hf.co/ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64
```

ruos-foundry-swarm-qwen3-30b-a3b-e64

ruos-foundry-swarm-qwen3-30b-a3b-e64 is a ruOS specialist cut from Qwen/Qwen3-30B-A3B-Instruct-2507 with MoE-Foundry (ADR-064): the parent's routed experts were traced on the ruOS ruos-swarm calibration set, the 64 most-used experts per routed layer were kept, and a complete smaller Mixture-of-Experts model was exported — same backbone, same tokenizer, same token top-k, fewer experts per layer. No weight was changed; experts were removed and renumbered and the router rows sliced to match.

Status: unevaluated and disabled

Per MoE-Foundry's rule an exported specialist starts quality_status: unevaluated, enabled: false. A structural export proves tensor integrity, not retained capability. This checkpoint is published so that slim-eval (slim/eval: run, redteam, regression, verdict) can measure it against the parent on the frozen ruOS test split; until that verdict is recorded here it must not be routed to. The full parent remains the fallback.

Parent


repo	`Qwen/Qwen3-30B-A3B-Instruct-2507`
revision	`0d7cf23991f47feeb3a57ecb4c9cee8ea4a17bfe`
licence	Apache-2.0
MoE-Foundry parent_id	`44c2b6971fe6979409ac72586db5fa4abfd73746afc52607368a59740848149d`
experts per routed layer (parent → this)	128 → 64
routed layers	48
token top-k (unchanged)	8

Calibration

Domain ruos-swarm: 64 rows (9 from the val split, 55 topped up from train; the test split is never used for calibration), ~22508 tokens, families {"swarm":64}, licences {"project-owned":64}. Calibration file sha256 94a5527b141269ab5078a14e3d129b387aa69412c65334160715fd9a58dbfef5. Texts are the ChatML prompt plus the reference answer.

Router traces: 412 tasks, 148144 tokens, 7110912 rows on NVIDIA A100-SXM4-80GB (bfloat16, transformers 4.51.3); trace sha256 7aabb05cbf715156437393fc5a5b0d756951ecee35eb4542a06fc812f11f9403.

Selection method: mass (accumulated routing probability per expert per layer) — a usage proxy, not causal importance.

Receipt


specialist_id	`cb102eec27bec133996063ed2999dbcd41f918a1f17fa212eac3271244fdeb93`
checkpoint_id	`66ea9f3a54b1e35812c8e8f429de84261e45670c66599e81f508f3a893f65eb4`
mask_sha256	`8edf60090f87097f9816e9e39fee82ef4c36ab24f3154867b155f89254de6210`
parent_id	`44c2b6971fe6979409ac72586db5fa4abfd73746afc52607368a59740848149d`
input tensor bytes	61064245248
output tensor bytes	32060633088
reduction	47.5%

separator_receipt.json in this repo is the full MoE-Foundry receipt including the per-layer expert mask.

Files

path	bytes	sha256
LICENSE	11343	05cab46843576551502bfdf712f84e93e6e9590d9997306ed4f6635ef82811d9
SHA256SUMS	737	7c0be9e4ae99b8d45f4b28a296f00d577cd05b645fd7a0684d2733abd54e0610
config.json	963	f0bf60e7da89b14b5763344441e4f60eccda09f62c0111c5f178283c5f3d81ec
generation_config.json	239	19d306dd769db12a9d710b44cf7f83b635efbe5166b84fb4358a08fb7d88bb53
merges.txt	1671839	599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3
model.safetensors	32061839304	4e75e1d20cba58fcf24d922eaa7a3b235dccd95c0ddcbe929238c5da7be029ac
separator_receipt.json	32971	643e8e09a99f60aaaf1737ef3bab4fdfc3bed4b12024f55201e58c93283ed33e
tokenizer.json	11422654	aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
tokenizer_config.json	9377	a62ff0a2472a0fa1b8eaabcb57c59b58afa42a22831dc141400b6e0cf2b65ce3
vocab.json	2776833	ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910

Run

vllm serve ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64 --max-model-len 8192

Loads with transformers as a standard qwen3_moe checkpoint (single safetensors file, num_experts reduced in config.json).

Limitations

Unevaluated: no capability, memory or latency claim is made here.
Retained experts were chosen by routing mass on ruOS calibration prompts; requests outside that domain should go to the parent.
Memory: fewer experts means a smaller checkpoint; loading several specialists next to the parent can use more total memory than the parent alone.

Provenance

MoE-Foundry 6677a25 (moe-separator inspect → profile_hf → select → export → mixture)
run foundry-20260907T172935Z-qwen3-30b-a3b on a single vast.ai GPU; ruos-desktop slim/foundry + slim/scripts/foundry-e2e.sh
authorisation: rUv, "implement this using ruvnet/MoE-Foundry using vast.ai in a worktree, implement e2e and push models to repo"

Downloads last month: 22

Safetensors

Model size

16B params

Tensor type

BF16

Model tree for ruvnet/ruos-foundry-swarm-qwen3-30b-a3b-e64

Base model

Qwen/Qwen3-30B-A3B-Instruct-2507

Finetuned

(125)

this model