Instructions to use dharun2049/Brahmaputra-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dharun2049/Brahmaputra-1 with Transformers:

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

pipe = pipeline("text-generation", model="dharun2049/Brahmaputra-1")
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("dharun2049/Brahmaputra-1")
model = AutoModelForCausalLM.from_pretrained("dharun2049/Brahmaputra-1", 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]:]))

PEFT
How to use dharun2049/Brahmaputra-1 with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dharun2049/Brahmaputra-1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dharun2049/Brahmaputra-1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dharun2049/Brahmaputra-1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dharun2049/Brahmaputra-1

SGLang

How to use dharun2049/Brahmaputra-1 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 "dharun2049/Brahmaputra-1" \
    --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": "dharun2049/Brahmaputra-1",
		"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 "dharun2049/Brahmaputra-1" \
        --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": "dharun2049/Brahmaputra-1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use dharun2049/Brahmaputra-1 with Docker Model Runner:
```
docker model run hf.co/dharun2049/Brahmaputra-1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Brahmaputra 1

Brahmaputra 1 starts from dharun2049/kaveri-stgrpo-0.5b and applies a compute-constrained contrastive knowledge-distillation stage using the dense Qwen/Qwen3.8-27B teacher.

Brahmaputra 1 is distilled from the starting checkpoint dharun2049/kaveri-stgrpo-0.5b using Qwen/Qwen3.8-27B as the teacher, with Codeforces anti-forgetting replay.

Teacher

Primary teacher: Qwen/Qwen3.8-27B

The distillation does not perform full-vocabulary KL. For each multiple-choice question, the teacher scores A/B/C/D and the student learns the teacher's four-way answer probability geometry.

Objective

L = 1.0 * L_MC + 1.5 * L_KD + 0.2 * L_margin + 0.35 * L_explanation

where:

L_MC is supervised four-way multiple-choice cross entropy.
L_KD is temperature-scaled KL divergence over A/B/C/D teacher scores.
L_margin forces the correct answer score above distractor scores.
L_explanation trains on selected concise teacher explanations and available native explanations.

Coding anti-forgetting

The CKD stage interleaves approximately 15% C++ replay micro-batches from open-r1/codeforces.

Before the fresh CKD LoRA is attached, the starting dharun2049/kaveri-stgrpo-0.5b checkpoint generates deterministic C++ replay trajectories. Coding batches then optimize:

L_code = 0.75 * L_replay_CE + 1.0 * L_anchor_KL

The KL reference is the exact pre-CKD Kaveri policy obtained by temporarily disabling the fresh CKD adapter. The anchor is computed over the reference policy's top-64 token support on completion positions.

Code-preservation probe before CKD: {'ce': 0.848503839224577, 'anchor_kl': -7.259872217280083e-10, 'n': 32}

Code-preservation probe after CKD: {'ce': 0.8353494852781296, 'anchor_kl': 0.0031774165108799934, 'n': 32}

Teacher KD is confidence-gated. If the teacher's top choice disagrees with the gold answer, that example receives zero KD weight, while its gold supervised loss remains active.

Data

MMLU dev/test were not loaded or used for training by the training script.

Training sources:

{ "medmcqa": 1631, "arc_challenge": 1032, "openbookqa": 1671, "arc_easy": 1510, "commonsenseqa": 1727 }

Teacher-scored examples: 7969

Qwen3.8-27B-generated explanations: 59

Training

Starting model: dharun2049/kaveri-stgrpo-0.5b
LoRA rank: 16
LoRA alpha: 32
LoRA targets: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
Optimizer steps completed: 279
Held-out external MCQ accuracy before CKD: 0.5251
Held-out external MCQ accuracy after CKD: 0.5276

The repository root contains the merged standalone model. The CKD LoRA adapter is preserved under ckd_adapter/.

Compute-budget note

The training script includes a wall-clock compute-unit governor. Colab does not expose a supported live CU billing API to Python, so the CU figure is an estimate based on a user-configurable assumed CU/hour rate.

Estimated CU at save: 7.10

Downloads last month: 112

Safetensors

Model size

0.5B params

Tensor type

BF16

Model tree for dharun2049/Brahmaputra-1

Base model

Qwen/Qwen2-0.5B

Finetuned

Qwen/Qwen2-0.5B-Instruct

Adapter

dharun2049/kaveri-stgrpo-0.5b

Adapter

(1)

this model

Quantizations

5 models