Gemma 3 27b it
Model Overview
Gemma 3 models are multimodal, handling text and image input and generating text output, with open weights for both pre-trained variants and instruction-tuned variants.
Gemma 3 models are well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as laptops, desktops or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone.
- Model Architecture: Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.
- Model Source: google/gemma-3-27b-it
- License: gemma
QPC Configuration # 1
| Precision |
SoCs / Tensor slicing |
NSP-Cores (per SoC) |
Full Batch Size |
Chunking Prompt Length |
Context Length (CL) |
QPC URL |
QPC Size |
QPC Download |
Onnx URL |
Onnx Download |
Generation Date |
| MXFP6 |
4 |
8 |
1 |
1024 |
16192 |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_qpc_8cores_1024pl_16192cl_4devices_prefill.tar.gz |
2.4GB |
Download |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_ONNX_prefill.tar.gz |
Download |
15-Apr-2026 |
| MXFP6 |
4 |
8 |
1 |
1024 |
16192 |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_qpc_8cores_1024pl_16192cl_4devices_decode.tar.gz |
44GB |
Download |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_ONNX_decode.tar.gz |
Download |
15-Apr-2026 |
QPC Configuration # 2
| Precision |
SoCs / Tensor slicing |
NSP-Cores (per SoC) |
Full Batch Size |
Chunking Prompt Length |
Context Length (CL) |
QPC URL |
QPC Size |
QPC Download |
Onnx URL |
Onnx Download |
Generation Date |
| MXFP6 |
2 |
16 |
1 |
128 |
4096 |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_qpc_16cores_128pl_4098cl_2devices_prefill.tar.gz |
1.1GB |
Download |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_ONNX_prefill.tar.gz |
Download |
15-Apr-2026 |
| MXFP6 |
2 |
16 |
1 |
128 |
4096 |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_qpc_16cores_128pl_4098cl_2devices_decode.tar.gz |
33GB |
Download |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_ONNX_decode.tar.gz |
Download |
15-Apr-2026 |
QPC Configuration # 3
| Precision |
SoCs / Tensor slicing |
NSP-Cores (per SoC) |
Full Batch Size |
Chunking Prompt Length |
Context Length (CL) |
QPC URL |
QPC Size |
QPC Download |
Onnx URL |
Onnx Download |
Generation Date |
| MXFP6 |
4 |
8 |
1 |
128 |
65536 |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_qpc_8cores_128pl_64kcl_[4k,8k,13k,16k,32k,64k]ccl_4device_prefill.tar.gz |
2.4GB |
Download |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_ONNX_prefill.tar.gz |
Download |
28-Apr-2026 |
| MXFP6 |
4 |
8 |
1 |
128 |
65536 |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_qpc_8cores_128pl_64kcl_[4k,8k,13k,16k,32k,64k]ccl_4devices_decode.tar.gz |
43GB |
Download |
https://dc00tk1pxen80.cloudfront.net/SDK1.21.4.0/google/gemma-3-27b-it/gemma-3-27b-it_ONNX_decode.tar.gz |
Download |
28-Apr-2026 |