Instructions to use AMbaye018/sarvam-30b-AWQ-then-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AMbaye018/sarvam-30b-AWQ-then-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="AMbaye018/sarvam-30b-AWQ-then-FP8", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("AMbaye018/sarvam-30b-AWQ-then-FP8", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AMbaye018/sarvam-30b-AWQ-then-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AMbaye018/sarvam-30b-AWQ-then-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AMbaye018/sarvam-30b-AWQ-then-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AMbaye018/sarvam-30b-AWQ-then-FP8

SGLang

How to use AMbaye018/sarvam-30b-AWQ-then-FP8 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 "AMbaye018/sarvam-30b-AWQ-then-FP8" \
    --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": "AMbaye018/sarvam-30b-AWQ-then-FP8",
		"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 "AMbaye018/sarvam-30b-AWQ-then-FP8" \
        --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": "AMbaye018/sarvam-30b-AWQ-then-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AMbaye018/sarvam-30b-AWQ-then-FP8 with Docker Model Runner:
```
docker model run hf.co/AMbaye018/sarvam-30b-AWQ-then-FP8
```

Sarvam-30B AWQ→FP8 (Mixed-Precision)

Hybrid mixed-precision quantization of sarvamai/sarvam-30b for the Resilient AI Challenge.

Method: AWQ first, then FP8

Two-stage sequential compression:

AWQ W4A16 on MLP/expert layers (4-bit, activation-aware scaling)
FP8 Dynamic on remaining BF16 layers (attention + layer 0)

This produces a hybrid model where each component uses the optimal precision:

Component	Precision	Why
MLP/Experts (layers 1-18)	INT4 (AWQ)	128 MoE experts tolerate 4-bit thanks to redundancy
Attention (layers 0-18)	FP8	Sensitive with only 4 KV heads, FP8 preserves quality
Layer 0 MLP (dense)	FP8	Dense layer (not MoE), more sensitive than experts
lm_head	BF16	Output layer, always kept at full precision

Quantization Details

Specification	Value
Method	AWQ W4A16 → FP8 Dynamic (sequential)
AWQ Tool	llm-compressor
AWQ Recipe	QuantTrio recipe (ignore attention + layer 0)
AWQ Calibration	sarvamai/indivibe + cais/mmlu
FP8 Scheme	FP8_DYNAMIC (no calibration needed)
Model Size	~24 GB (vs 60 GB baseline, 26 GB AWQ-only, 37 GB FP8-only)
Hardware	Quantized on NVIDIA H100 80GB

Usage with vLLM

vllm serve AMbaye018/sarvam-30b-AWQ-then-FP8 \
    --trust-remote-code \
    --tensor-parallel-size 1 \
    --gpu-memory-utilization 0.90 \
    --max-model-len 32768 \
    --host 0.0.0.0 \
    --port 8000

vLLM Config (for challenge submission)

model: AMbaye018/sarvam-30b-AWQ-then-FP8
served_model_name: sarvam-30b-awq-fp8
trust_remote_code: true
tensor_parallel_size: 1
gpu_memory_utilization: 0.90
max_model_len: 32768
max_num_seqs: 64
host: 0.0.0.0
port: 8000

References

sarvamai/sarvam-30b — Base model
QuantTrio/sarvam-30b-AWQ — AWQ recipe reference
llm-compressor — Quantization tool

License

Apache License 2.0 (same as base model)

Downloads last month: 2

Safetensors

Model size

7B params

Tensor type

I64

F32

I32

F8_E4M3

Model tree for AMbaye018/sarvam-30b-AWQ-then-FP8

Base model

sarvamai/sarvam-30b

Quantized

(21)

this model