Skip to content

Gpt oss 120b

Model Overview

OpenAI’s GPT-OSS models (gpt-oss-120b & gpt-oss-20b) are open-weight models designed for powerful reasoning, agentic tasks and versatile developer use cases. GPT-OSS-120B is used for production, general purpose, high reasoning use cases that fit into a single 80GB GPU.

  • Model Architecture: 117B parameters with 5.1B active parameters. Trained on harmony response format and should only be used with the harmony format as it will not work correctly otherwise.
  • Model Source: openai/gpt-oss-120b
  • License: Apache 2.0 license. Build freely without copyleft restrictions or patent risk—ideal for experimentation, customization, and commercial deployment.
  • Configurable reasoning effort: Easily adjust the reasoning effort (low, medium, high) based on your specific use case and latency needs.
  • Full chain-of-thought: Gain complete access to the model’s reasoning process, facilitating easier debugging and increased trust in outputs. It’s not intended to be shown to end users.
  • Fine-tunable: Fully customize models to your specific use case through parameter fine-tuning.
  • Agentic capabilities: Use the models’ native capabilities for function calling, web browsing, Python code execution, and Structured Outputs.
  • Native MXFP4 quantization: The models were post-trained with MXFP4 quantization of the MoE weights, making gpt-oss-120b run on a single 80GB GPU.

QPC Configuration - 1

Precision SoCs / Tensor slicing NSP-Cores (per SoC) Batch Size Chunking Prompt Length Context Length (CL) Generated URL Download Generation Date
MXFP6 4 16 1 256 8192 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_8192cl_prefill.tar.gz Download 13-Feb-2026
MXFP6 4 16 1 256 8192 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_8192cl_decode.tar.gz Download 13-Feb-2026

QPC Configuration - 2

Precision SoCs / Tensor slicing NSP-Cores (per SoC) Batch Size Chunking Prompt Length Context Length (CL) Generated URL Download Generation Date
MXFP6 4 16 1 128 16384 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_128pl_16384cl_prefill.tar.gz Download 13-Feb-2026
MXFP6 4 16 1 128 16384 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_128pl_16384cl_decode.tar.gz Download 13-Feb-2026

QPC Configuration - 3

Precision SoCs / Tensor slicing NSP-Cores (per SoC) Batch Size Chunking Prompt Length Context Length (CL) Generated URL Download Generation Date
MXFP6 4 16 1 256 32768 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_32kcl_prefill.tar.gz Download 16-Feb-2026
MXFP6 4 16 1 256 32768 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_32kcl_decode.tar.gz Download 16-Feb-2026

QPC Configuration - 4

Precision SoCs / Tensor slicing NSP-Cores (per SoC) Batch Size Chunking Prompt Length Context Length (CL) Generated URL Download Generation Date
MXFP6 4 16 1 256 16384 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_16384cl_prefill.tar.gz Download 18-Feb-2026
MXFP6 4 16 1 256 16384 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_16384cl_decode.tar.gz Download 18-Feb-2026

QPC Configuration - 5

Precision SoCs / Tensor slicing NSP-Cores (per SoC) Batch Size Chunking Prompt Length Context Length (CL) Generated URL Download Generation Date
MXFP6 4 16 1 256 13312 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_13312cl_prefill.tar.gz Download 19-Feb-2026
MXFP6 4 16 1 256 13312 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_13312cl_decode.tar.gz Download 19-Feb-2026

QPC Configuration - 6

Precision SoCs / Tensor slicing NSP-Cores (per SoC) Batch Size Chunking Prompt Length Context Length (CL) Generated URL Download Generation Date
MXFP6 4 16 1 256 4096 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_4096cl_prefill.tar.gz Download 19-Feb-2026
MXFP6 4 16 1 256 4096 https://qualcom-qpc-models.s3-accelerate.amazonaws.com/SDK1.21A1.1/openai/gpt-oss-120b/gpt_oss120b_qpc_256pl_4096cl_decode.tar.gz Download 19-Feb-2026

Run This Model

Download QPCs

mkdir -p openai/gpt-oss-120b
cd openai/gpt-oss-120b

# Download Prefill QPC
wget <Prefill_QPC_Download_URL>
tar xzvf <prefill_qpc_filename.tar.gz>

# Download Decode QPC
wget <Decode_QPC_Download_URL>
tar xzvf <decode_qpc_filename.tar.gz>

Run QPC

Replace prefill_qpc_path and decode_qpc_path with actual extracted QPC directories.

python3 -m qaic_disagg \
  --prefill-port 9800 \
  --decode-port 9900 \
  --port 8082 \
  --decode-device-group 4,5,6,7 \
  --prefill-device-group 0,1,2,3 \
  --model openai/gpt-oss-120b \
  --prefill-max-num-seqs 1 \
  --decode-max-num-seqs 1 \
  --prefill-max-seq-len-to-capture <128/256> \
  --max-model-len <context_length(cl)> \
  --prefill-override-qaic-config "split_retained_state_io:True mxfp6_matmul:True enable_chunking:True node_precision_info=/workspace/examples/disagg_serving/non_subfunction_120b_npi.yaml qpc_path=<prefill_qpc_path>" \
  --decode-override-qaic-config "mxfp6_matmul:True retain_full_kv:True node_precision_info=/workspace/examples/disagg_serving/non_subfunction_120b_npi.yaml qpc_path=<decode_qpc_path>" \
  -vvv \
  --dtype bfloat16 \
  --kv-cache-dtype mxint8 \
  --kv-handOff-port 5066 \
  --tool-call-parser openai \
  --enable-auto-tool-choice \
  --enable-log-outputs \
  --reasoning-parser openai_gptoss \
  --generation_config vllm \
  --enable-log-requests \
  --chat-template-content-format openai