Instructions to use webbrain-one/webbrain-compass-tiny-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use webbrain-one/webbrain-compass-tiny-v2 with Transformers:

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

pipe = pipeline("text-generation", model="webbrain-one/webbrain-compass-tiny-v2")
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("webbrain-one/webbrain-compass-tiny-v2")
model = AutoModelForCausalLM.from_pretrained("webbrain-one/webbrain-compass-tiny-v2", 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 webbrain-one/webbrain-compass-tiny-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "webbrain-one/webbrain-compass-tiny-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "webbrain-one/webbrain-compass-tiny-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/webbrain-one/webbrain-compass-tiny-v2

SGLang

How to use webbrain-one/webbrain-compass-tiny-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 "webbrain-one/webbrain-compass-tiny-v2" \
    --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": "webbrain-one/webbrain-compass-tiny-v2",
		"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 "webbrain-one/webbrain-compass-tiny-v2" \
        --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": "webbrain-one/webbrain-compass-tiny-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use webbrain-one/webbrain-compass-tiny-v2 with Docker Model Runner:
```
docker model run hf.co/webbrain-one/webbrain-compass-tiny-v2
```

webbrain-compass-tiny-v2 / provenance /candidate-verification.json

webbrain-one

Upload private Compass v2 model batch 2/3

a9d1f19 verified 30 days ago

Raw History Blame Contribute Delete

15.8 kB

{"comparisons": {"merged_fp32__vs__merged_bf16": [{"greedyIdentical": true, "maxAbsolute": 0.564452052116394, "maxKL": 0.00718469824641943, "meanKL": 0.0015943044563755393, "probe": "brief_thanks", "relativeL2": 0.027548564597964287, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.5689697265625, "maxKL": 0.004700065590441227, "meanKL": 0.0008369088172912598, "probe": "definition_tab", "relativeL2": 0.024597246199846268, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.6966772079467773, "maxKL": 0.004516322631388903, "meanKL": 0.0009982013143599033, "probe": "turkish_sentence", "relativeL2": 0.03252628818154335, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.612119197845459, "maxKL": 0.004691502545028925, "meanKL": 0.0011473649647086859, "probe": "plain_summary", "relativeL2": 0.02769998274743557, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 1.0917444229125977, "maxKL": 0.00644421624019742, "meanKL": 0.0010239672847092152, "probe": "single_address", "relativeL2": 0.02633913792669773, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.7655773162841797, "maxKL": 0.006034428719431162, "meanKL": 0.0008162062731571496, "probe": "escaped_address", "relativeL2": 0.023054610937833786, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.9427709579467773, "maxKL": 0.010114249773323536, "meanKL": 0.0014173928648233414, "probe": "typed_labels", "relativeL2": 0.025686681270599365, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.7821457386016846, "maxKL": 0.01134589221328497, "meanKL": 0.001547513296827674, "probe": "tool_history", "relativeL2": 0.024425603449344635, "top1Agreement": 1.0}], "unmerged_bf16__vs__merged_bf16": [{"greedyIdentical": true, "maxAbsolute": 0.9375, "maxKL": 0.008973976597189903, "meanKL": 0.0022216755896806717, "probe": "brief_thanks", "relativeL2": 0.03687362000346184, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.75, "maxKL": 0.017957624047994614, "meanKL": 0.0024153669364750385, "probe": "definition_tab", "relativeL2": 0.03703984618186951, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.890625, "maxKL": 0.007195579819381237, "meanKL": 0.0012555174762383103, "probe": "turkish_sentence", "relativeL2": 0.041932184249162674, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.8984375, "maxKL": 0.01066482625901699, "meanKL": 0.0026590011548250914, "probe": "plain_summary", "relativeL2": 0.03822087496519089, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 1.0234375, "maxKL": 0.00833512470126152, "meanKL": 0.001597078051418066, "probe": "single_address", "relativeL2": 0.03441133722662926, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.89453125, "maxKL": 0.012166930362582207, "meanKL": 0.001568951760418713, "probe": "escaped_address", "relativeL2": 0.03155586123466492, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.875, "maxKL": 0.04109187424182892, "meanKL": 0.005159610416740179, "probe": "typed_labels", "relativeL2": 0.03392578288912773, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 1.015625, "maxKL": 0.012617651373147964, "meanKL": 0.0022947993129491806, "probe": "tool_history", "relativeL2": 0.030292633920907974, "top1Agreement": 1.0}], "unmerged_fp32__vs__unmerged_bf16": [{"greedyIdentical": true, "maxAbsolute": 0.6285281181335449, "maxKL": 0.007544202730059624, "meanKL": 0.0016296862158924341, "probe": "brief_thanks", "relativeL2": 0.029022900387644768, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.6977815628051758, "maxKL": 0.015743080526590347, "meanKL": 0.0021585384383797646, "probe": "definition_tab", "relativeL2": 0.029428735375404358, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.7233128547668457, "maxKL": 0.006144802551716566, "meanKL": 0.0013706989120692015, "probe": "turkish_sentence", "relativeL2": 0.031044388189911842, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.7862186431884766, "maxKL": 0.007198987528681755, "meanKL": 0.0014474603813141584, "probe": "plain_summary", "relativeL2": 0.03223147988319397, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 1.0966455936431885, "maxKL": 0.004912373144179583, "meanKL": 0.0010347821516916156, "probe": "single_address", "relativeL2": 0.030492622405290604, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 1.0286784172058105, "maxKL": 0.00555111700668931, "meanKL": 0.0007849412504583597, "probe": "escaped_address", "relativeL2": 0.029800396412611008, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.6691875457763672, "maxKL": 0.03257470205426216, "meanKL": 0.004195042420178652, "probe": "typed_labels", "relativeL2": 0.026134153828024864, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.810427188873291, "maxKL": 0.003376855980604887, "meanKL": 0.0006764935096725821, "probe": "tool_history", "relativeL2": 0.024038691073656082, "top1Agreement": 1.0}]}, "contractSha256": "81675a38e6d3967a80a59b10787cb4356536b96e38f7630c7245ae1cd2832f5b", "files": [{"bytes": 9060, "path": "chat_template.jinja", "sha256": "cc945752db555d60949b16989df4ccfeb52a313d6b4b5c5229dd786e2e9fcf1c"}, {"bytes": 784, "path": "config.json", "sha256": "ef8d508998a15182d9295485e83537c8128b86dab2a7859370ce3f3cff0e4f26"}, {"bytes": 186, "path": "generation_config.json", "sha256": "6f66aa5aaf68145027c3f3c384abfd837530fe4d1efb582e0aa919eecd4c9a8e"}, {"bytes": 3995365288, "path": "model-00001-of-00002.safetensors", "sha256": "7c1987b5447a28eb20bfd3735b2d63e4d5eace5bf143980e72306727ee318a8f"}, {"bytes": 1038191672, "path": "model-00002-of-00002.safetensors", "sha256": "a81767b62535403421571c90b1abe4f907101676133d2136407fd5ee35046314"}, {"bytes": 31804, "path": "model.safetensors.index.json", "sha256": "8aa878d625668a709f44ddca54d4643781d8cfc6f6cc5aa85363128aaabef32c"}, {"bytes": 9894271, "path": "tokenizer.json", "sha256": "3e065a558a034185fe299917b398685c1facd0169a9eea1e629eb30c171fed81"}, {"bytes": 451, "path": "tokenizer_config.json", "sha256": "9ade1bc52582b9678995023a08617ab19528347811d06cb4ed658bc6d9348823"}], "fp32": [{"greedyIdentical": true, "maxAbsolute": 4.601478576660156e-05, "maxKL": 1.618886180665413e-08, "meanKL": -3.503504686364067e-08, "probe": "brief_thanks", "relativeL2": 2.4299492906720843e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 6.103515625e-05, "maxKL": 8.317044830619125e-08, "meanKL": 1.980265906809109e-08, "probe": "definition_tab", "relativeL2": 2.504981011952623e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 6.532669067382812e-05, "maxKL": 8.356306580026285e-08, "meanKL": -2.1137747552302244e-08, "probe": "turkish_sentence", "relativeL2": 3.1571482850267785e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 5.9723854064941406e-05, "maxKL": 1.6377316569560207e-07, "meanKL": 1.775125646474862e-08, "probe": "plain_summary", "relativeL2": 2.8717195164063014e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 6.580352783203125e-05, "maxKL": 3.403988557693083e-07, "meanKL": 3.717636332112306e-08, "probe": "single_address", "relativeL2": 2.4546995973651065e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.00010204315185546875, "maxKL": 2.524157594052667e-07, "meanKL": 4.863847280489608e-08, "probe": "escaped_address", "relativeL2": 2.8070771804777905e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 6.67572021484375e-05, "maxKL": 2.8021582920700894e-07, "meanKL": 4.0257429390067045e-08, "probe": "typed_labels", "relativeL2": 2.6742122827272397e-06, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 8.106231689453125e-05, "maxKL": 2.3862357423354297e-08, "meanKL": -1.4172083950825254e-08, "probe": "tool_history", "relativeL2": 2.947297161881579e-06, "top1Agreement": 1.0}], "implementationSha256": "8ff37dca66f7d9f97d7e02723f12d5d0da8b3c4ece16aefec3bbe85695427af3", "independentWeightCheckSha256": "8488511233d9803036b554e45c10adc8e0092a87f7eb24e2643bf8b3720fef01", "kv": {"merged_bf16": [{"cacheLength": 26, "decodeLogitsSha256": "e33699b4a7ea6c5f1893ba35187f3b170f70160114685e132084af50c282fd97", "decodedToken": 18020, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.375, "nextTop1": 447, "prefillTokens": 25, "probe": "brief_thanks"}, {"cacheLength": 24, "decodeLogitsSha256": "2b748266c87d84156319e8a7e521e54e395013b04961a3360332dc6c939f746b", "decodedToken": 54, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.1640625, "nextTop1": 28511, "prefillTokens": 23, "probe": "definition_tab"}, {"cacheLength": 38, "decodeLogitsSha256": "7551617564c92438e6629840debcb992b9b6616c1892512f6ca48a5cc8240c01", "decodedToken": 87776, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.25, "nextTop1": 7320, "prefillTokens": 37, "probe": "turkish_sentence"}, {"cacheLength": 31, "decodeLogitsSha256": "822c0c8b0cffcf505ffe0a853768496bbcc2c21595fd4a6e863b1efc5d70219a", "decodedToken": 5615, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.2578125, "nextTop1": 317, "prefillTokens": 30, "probe": "plain_summary"}, {"cacheLength": 222, "decodeLogitsSha256": "66bac405f2023eb8055888f372057cf3d0e3d5cabeb423491ff109a67c9f379b", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.40625, "nextTop1": 2546, "prefillTokens": 221, "probe": "single_address"}, {"cacheLength": 229, "decodeLogitsSha256": "ca2b42a5cf5474085e8ab7d77211793eaaa91315a4afb6f9d6db3f9cc194d142", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.390625, "nextTop1": 2546, "prefillTokens": 228, "probe": "escaped_address"}, {"cacheLength": 245, "decodeLogitsSha256": "7d0f1334e416460f07c1cc387c5b87f5d7204e6c6e8c33f430a6b6b456368f04", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.484375, "nextTop1": 2546, "prefillTokens": 244, "probe": "typed_labels"}, {"cacheLength": 303, "decodeLogitsSha256": "dd5e7ee4acb0c021283630072caf07eed28a775a8a1a9f21c69a8170bc8669ad", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.46875, "nextTop1": 2546, "prefillTokens": 302, "probe": "tool_history"}], "merged_fp32": [], "reloaded_bf16": [{"cacheLength": 26, "decodeLogitsSha256": "e33699b4a7ea6c5f1893ba35187f3b170f70160114685e132084af50c282fd97", "decodedToken": 18020, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.375, "nextTop1": 447, "prefillTokens": 25, "probe": "brief_thanks"}, {"cacheLength": 24, "decodeLogitsSha256": "2b748266c87d84156319e8a7e521e54e395013b04961a3360332dc6c939f746b", "decodedToken": 54, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.1640625, "nextTop1": 28511, "prefillTokens": 23, "probe": "definition_tab"}, {"cacheLength": 38, "decodeLogitsSha256": "7551617564c92438e6629840debcb992b9b6616c1892512f6ca48a5cc8240c01", "decodedToken": 87776, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.25, "nextTop1": 7320, "prefillTokens": 37, "probe": "turkish_sentence"}, {"cacheLength": 31, "decodeLogitsSha256": "822c0c8b0cffcf505ffe0a853768496bbcc2c21595fd4a6e863b1efc5d70219a", "decodedToken": 5615, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.2578125, "nextTop1": 317, "prefillTokens": 30, "probe": "plain_summary"}, {"cacheLength": 222, "decodeLogitsSha256": "66bac405f2023eb8055888f372057cf3d0e3d5cabeb423491ff109a67c9f379b", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.40625, "nextTop1": 2546, "prefillTokens": 221, "probe": "single_address"}, {"cacheLength": 229, "decodeLogitsSha256": "ca2b42a5cf5474085e8ab7d77211793eaaa91315a4afb6f9d6db3f9cc194d142", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.390625, "nextTop1": 2546, "prefillTokens": 228, "probe": "escaped_address"}, {"cacheLength": 245, "decodeLogitsSha256": "7d0f1334e416460f07c1cc387c5b87f5d7204e6c6e8c33f430a6b6b456368f04", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.484375, "nextTop1": 2546, "prefillTokens": 244, "probe": "typed_labels"}, {"cacheLength": 303, "decodeLogitsSha256": "dd5e7ee4acb0c021283630072caf07eed28a775a8a1a9f21c69a8170bc8669ad", "decodedToken": 18, "fullRecomputeTop1Identical": true, "maxAbsolute": 0.46875, "nextTop1": 2546, "prefillTokens": 302, "probe": "tool_history"}], "unmerged_bf16": [], "unmerged_fp32": []}, "modelDir": "E:\\webbrain-curated-v2-sources\\work\\curated-v2\\compass-v2-model-release-v1\\compass-v2-bf16-v4_2-805c995d3797b8a3\\model", "oldV1V2GateStatus": "failed-preserved", "releaseApproved": false, "reload": [{"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "brief_thanks", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "definition_tab", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "turkish_sentence", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "plain_summary", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "single_address", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "escaped_address", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "typed_labels", "relativeL2": 0.0, "top1Agreement": 1.0}, {"greedyIdentical": true, "maxAbsolute": 0.0, "maxKL": 0.0, "meanKL": 0.0, "probe": "tool_history", "relativeL2": 0.0, "top1Agreement": 1.0}], "source": {"adapterSha256": "2e1d86e4fd7c377a2db33bf93af6951bf7db941d0a265eed8a934a6fc42feefe", "completionSha256": "805c995d3797b8a3f7ddd402ccf84773021affcafe8a206086ca2dfe06b3c31d", "training": {"configSha256": "0f17b59149993b54d37d742c38498c2866edcdaa24e545300b0f4b7ea7302144", "datasetManifestSha256": "822a167383353f3942f59ad042363fd3bd7b8f5df767b0338345d861cbcca4b1", "finalValidation": {"epoch": 1.0, "eval_ask_loss": 1.2844717502593994, "eval_ask_runtime": 31.426, "eval_ask_samples_per_second": 1.814, "eval_ask_steps_per_second": 1.814, "eval_compact-action-noncommercial-weblinx_loss": 0.011626693420112133, "eval_compact-action-noncommercial-weblinx_runtime": 28.5725, "eval_compact-action-noncommercial-weblinx_samples_per_second": 1.4, "eval_compact-action-noncommercial-weblinx_steps_per_second": 1.4, "eval_compact-action_loss": 1.1079974174499512, "eval_compact-action_runtime": 96.4125, "eval_compact-action_samples_per_second": 1.151, "eval_compact-action_steps_per_second": 1.151, "eval_compact-escalation_loss": 1.4954874515533447, "eval_compact-escalation_runtime": 3.405, "eval_compact-escalation_samples_per_second": 1.468, "eval_compact-escalation_steps_per_second": 1.468, "num_input_tokens_seen": 76993328}, "optimizerStep": 457, "optimizerStepsTotal": 457, "resumeFromCheckpoint": null, "schemaVersion": 1, "sealedTestOpened": false, "smokeSteps": 0, "stateLedger": "E:\\webbrain-curated-v2-sources\\runs\\webbrain-compass-v2-minicpm5-2b-reviewed-runtime-v2\\webbrain-compass-v2-minicpm5-2b-sft-v1-b3c9b2327bfbd41a\\full-one-epoch\\training-state.append-only.jsonl", "status": "webbrain-compass-v2-training-complete", "trainableParameters": 25116672, "trainingLoss": 0.893471076921509}, "trainingReportSha256": "31e459f53bd548e82e8930596107941ccf23c1de2290446be037d48e05d0836c", "verifiedArtifacts": 153}, "status": "candidate-numerical-reload-verified-benchmark-pending", "weightBitInventorySha256": "c1b79bb94f5f33f14fdabaf03c689ef04b26de1c3ea59a46242f9e7b1b0d0db6"}