Sarvam m
Model Overview¶
sarvam-m is a multilingual, hybrid-reasoning, text-only language model built on Mistral-Small. This post-trained version delivers exceptional improvements over the base model: +20% average improvement on Indian language benchmarks, +21.6% enhancement on math benchmarks, +17.6% boost on programming benchmarks.
Key Features include Hybrid Thinking Mode, Advanced Indic Skills, Superior Reasoning Capabilities & Seamless Chatting Experience.
Model Architecture¶
- Model Source: sarvamai/sarvam-m
- License: licenses/markdown/apache-2.0.md
QPC Configurations¶
| Precision | SoCs / Tensor slicing | NSP-Cores (per SoC) | Full Batch Size | Chunking Prompt Length | Context Length (CL) | Generated URL | Download | Generation Date |
|---|---|---|---|---|---|---|---|---|
| MXFP6 | 4 | 16 | 1 | 128 | 8192 | https://dc00tk1pxen80.cloudfront.net/SDK1.20.4/sarvamai/sarvam-m/sarvam-m_qpc_16cores_128pl_8192cl_1fbs_4devices_mxfp6_mxint8.tar.gz | Download | 27-Feb-2026 |
| MXFP6 | 8 | 16 | 1 | 128 | 8192 | https://dc00tk1pxen80.cloudfront.net/SDK1.20.4/sarvamai/sarvam-m/sarvam-m_qpc_16cores_128pl_8192cl_1fbs_8devices_mxfp6_mxint8.tar.gz | Download | 27-Feb-2026 |
Run This Model¶
# Download QPC
mkdir -p sarvamai/sarvam-m
cd sarvamai/sarvam-m
wget <Download URL>
tar xzvf <downloaded filename.tar.gz>
# Run QPC
python3 -m QEfficient.cloud.execute --model_name sarvamai/sarvam-m --qpc_path <path/to/qpc> --prompt "# shortest path algorithm\n" --generation_len 128
API Endpoint¶
# Start REST endpoint with vLLM
VLLM_QAIC_MAX_CPU_THREADS=8 VLLM_QAIC_QPC_PATH=/path/to/qpc python3 -m vllm.entrypoints.openai.api_server \
--host 0.0.0.0 \
--port 8000 \
--model sarvamai/sarvam-m \
--max-model-len <Context Length> \
--max-num-seq <Full Batch Size> \
--max-seq_len-to-capture <Chunking Prompt Length> \
--device qaic \
--block-size 32