Mistral 7B Instruct v0.1
Model Overview¶
The Mistral-7B-Instruct-v0.1 Large Language Model (LLM) is a instruct fine-tuned version of the Mistral-7B-v0.1 generative text model using a variety of publicly available conversation datasets.
-
Paper: Mistral 7B
-
Model Source: mistralai/Mistral-7B-Instruct-v0.1
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 | 16 | 128 | 4096 | https://dc00tk1pxen80.cloudfront.net/SDK1.20.4/mistralai/Mistral-7B-Instruct-v0.1/qpc_16cores_128pl_4096cl_16fbs_4devices_mxfp6_mxint8.tar.gz | Download | |
| MXFP6 | 4 | 16 | 1 | 128 | 4096 | https://dc00tk1pxen80.cloudfront.net/SDK1.20.4/mistralai/Mistral-7B-Instruct-v0.1/qpc_16cores_128pl_4096cl_1fbs_4devices_mxfp6_mxint8.tar.gz | Download | |
| MXFP6 | 4 | 16 | 8 | 128 | 4096 | https://dc00tk1pxen80.cloudfront.net/SDK1.20.4/mistralai/Mistral-7B-Instruct-v0.1/qpc_16cores_128pl_4096cl_8fbs_4devices_mxfp6_mxint8.tar.gz | Download | |
| MXFP6 | 2 | 16 | 1 | 128 | 4096 | https://dc00tk1pxen80.cloudfront.net/SDK1.20.4/mistralai/Mistral-7B-Instruct-v0.1/Mistral-7B-Instruct-v0.1_qpc_16cores_128pl_4096cl_1fbs_2devices_mxfp6_mxint8.tar.gz | Download | 21-Jan-2026 |
Run This Model¶
# Download QPC
mkdir -p mistralai/Mistral-7B-Instruct-v0.1
cd mistralai/Mistral-7B-Instruct-v0.1
wget <Download URL>
tar xzvf <downloaded filename.tar.gz>
# Run QPC
python3 -m QEfficient.cloud.execute --model_name mistralai/Mistral-7B-Instruct-v0.1 --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 mistralai/Mistral-7B-Instruct-v0.1 \
--max-model-len <Context Length> \
--max-num-seq <Full Batch Size> \
--max-seq_len-to-capture <Chunking Prompt Length> \
--device qaic \
--block-size 32