Instructions to use AxisMeru/prabhasa-nyaya-element-judge-4b-arm0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AxisMeru/prabhasa-nyaya-element-judge-4b-arm0 with Transformers:

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

pipe = pipeline("text-generation", model="AxisMeru/prabhasa-nyaya-element-judge-4b-arm0")
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("AxisMeru/prabhasa-nyaya-element-judge-4b-arm0")
model = AutoModelForCausalLM.from_pretrained("AxisMeru/prabhasa-nyaya-element-judge-4b-arm0", 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 AxisMeru/prabhasa-nyaya-element-judge-4b-arm0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AxisMeru/prabhasa-nyaya-element-judge-4b-arm0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AxisMeru/prabhasa-nyaya-element-judge-4b-arm0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AxisMeru/prabhasa-nyaya-element-judge-4b-arm0

SGLang

How to use AxisMeru/prabhasa-nyaya-element-judge-4b-arm0 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 "AxisMeru/prabhasa-nyaya-element-judge-4b-arm0" \
    --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": "AxisMeru/prabhasa-nyaya-element-judge-4b-arm0",
		"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 "AxisMeru/prabhasa-nyaya-element-judge-4b-arm0" \
        --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": "AxisMeru/prabhasa-nyaya-element-judge-4b-arm0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AxisMeru/prabhasa-nyaya-element-judge-4b-arm0 with Docker Model Runner:
```
docker model run hf.co/AxisMeru/prabhasa-nyaya-element-judge-4b-arm0
```

prabhasa-nyaya-element-judge-4b-arm0

Part of Prabhāsa-Nyāya, our own trained Indian-legal models; used by the Pravrudhi-Nyaya product.

Deployed as production 4B judge 2026-09-30; serving integrity verified; quality claims at constructed tier only. EXPLORATORY. Single seed.

Scope of the 0/225 figure: Constructed in-distribution set, two-judge config C; not real court material. On real court judgments (single-judge, 1 Oct): 11/122 false proofs, CP95 upper 14.5%.

Full fine-tune of Qwen/Qwen3-4B as an element judge for the Nyaya pipeline ("is this legal element ESTABLISHED by these facts?"). Output is a short verdict whose first-token logprobs give p_established.

Training

Data: element_judgment_combined.jsonl @36a69ec, SHA-256 53697c5762c011b57cf80e00d9ddf6aa5cb54eae948eca8b59c569d7123ab440, 1,463 rows (2,074 balanced train / 136 val). Same file as the deployed 4B v0.
Completion-only loss (completion_only_loss=True): prompt tokens are masked, only the verdict completion is trained. This is the only intended difference from the deployed 4B, which trained on all tokens. (Masking check: arm0_masking_evidence.json.)
Base: Qwen/Qwen3-4B, full fine-tune (no LoRA). Base revision 1cfa9a72… (1cfa9a7208912126459214e8b04321603b3df60c) is inferred: the training run loaded the unpinned name, this was the only snapshot in the training machine's cache, and the trainer did not record it. The base is not provably pinned.
Hyperparameters: 3 epochs (~390 optimizer steps), batch 4, grad-accum 4, lr 2e-5 cosine, warmup 100, adamw_bnb_8bit, gradient checkpointing, bf16, max_length 1024, seed default. Identical to the deployed 4B's recipe apart from the loss mask.
TRL 1.13.0, Transformers 5.17.0, PyTorch 2.9.0a0 (NVIDIA 25.09), Datasets 5.0.1, Tokenizers 0.23.2. bf16, single seed.
Weights file model.safetensors SHA-256 c2fa5954d5bab3885d13d472e38aad67fe6cbad51f54a08a472fa2f9a6a9eec6.

Results (all from sealed, committed runs; caveats bind every number)

Element level, tau = 0.74, Arm-0 vs the deployed 4B, exact paired McNemar (two-sided):

Established-side (est200, original fact order): 200/200 vs 183/200 (Arm-0-only 17, deployed-only 0, p 1.5e-5).
NE-side (ne71 wrong-ESTABLISHED, original order): 6/71 vs 12/71, not significant (p 0.109). Do not read this as Arm-0 beating the deployed 4B on not-established.
Shuffled fact order on ne71: 7/71 vs 23/71 wrong-ESTABLISHED (p 1.45e-4).
Arm-0 (completion-only 4B): DEPLOYED to production 2026-09-30 (serving integrity verified: served weights reproduce sealed scores). Config-C on the constructed frozen sets, production fact-id convention: 117/377 Lean-verified proofs vs deployed 105/377 on the same fresh path (12 vs 0 discordant, one-sided exact p=0.00024). Against the earlier 109/377 replay (frozen-logit F_{element_id} convention, not production): 8 vs 0, p=0.0039. False proofs 0/225 all legs (not fresh evidence). Seal-order slip disclosed (log printed before commit); outputs byte-identical on R1's independent re-run. Tier: constructed, in-distribution, single seed. No claim of fewer wrong establishments; transfer to real material untested.
Config-C = AND of this 4B at tau 0.74, the frozen 32B judge at 0.97, and the Lean registry check (a structural check, not proof verification).

Limitations

The 377/225/271 evaluation sets are disjoint from the training file by id and text, but in-distribution: same task family, same generators. No transfer or real-material claim.
One seed, one shuffle, no multiplicity correction; units within a contract are not independent.
The Lean step in config-C is a structural check, not a proof verification.
Not evaluated for use outside this pipeline's prompt format; serving requires the pinned prompt builder and top_logprobs.
Transfer to other data is untested; no not-established superiority claim.
Serving-switch staging and a rollback path are pre-registered separately; this repo is the artifact only.

Files

Weights, config, tokenizer and chat template as trained. training_args.bin is the TRL arguments pickle; do not load it untrusted.

Downloads last month: 1,166

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for AxisMeru/prabhasa-nyaya-element-judge-4b-arm0

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Finetuned

(1190)

this model

Adapters

1 model