cpu.md 1.86 KB
Newer Older
Louie Tsai's avatar
Louie Tsai committed
1
2
# CPU - Intel® Xeon®

3
4
## Validated Hardware

5
6
7
8
| Hardware |
| -------- |
| [Intel® Xeon® 6 Processors](https://www.intel.com/content/www/us/en/products/details/processors/xeon.html) |
| [Intel® Xeon® 5 Processors](https://www.intel.com/content/www/us/en/products/docs/processors/xeon/5th-gen-xeon-scalable-processors.html) |
9

10
## Recommended Models
Louie Tsai's avatar
Louie Tsai committed
11
12
13
14

### Text-only Language Models

| Model                                | Architecture                             | Supported |
15
16
17
18
19
20
21
22
23
| ------------------------------------ | ---------------------------------------- | --------- |
| meta-llama/Llama-3.1-8B-Instruct     | LlamaForCausalLM                         | ✅        |
| meta-llama/Llama-3.2-3B-Instruct     | LlamaForCausalLM                         | ✅        |
| ibm-granite/granite-3.2-2b-instruct  | GraniteForCausalLM                       | ✅        |
| Qwen/Qwen3-1.7B                      | Qwen3ForCausalLM                         | ✅        |
| Qwen/Qwen3-4B                        | Qwen3ForCausalLM                         | ✅        |
| Qwen/Qwen3-8B                        | Qwen3ForCausalLM                         | ✅        |
| zai-org/glm-4-9b-hf                  | GLMForCausalLM                           | ✅        |
| google/gemma-7b                      | GemmaForCausalLM                         | ✅        |
Louie Tsai's avatar
Louie Tsai committed
24
25
26
27

### Multimodal Language Models

| Model                                | Architecture                             | Supported |
28
29
30
| ------------------------------------ | ---------------------------------------- | --------- |
| Qwen/Qwen2.5-VL-7B-Instruct          | Qwen2VLForConditionalGeneration          | ✅        |
| openai/whisper-large-v3              | WhisperForConditionalGeneration          | ✅        |
Louie Tsai's avatar
Louie Tsai committed
31
32
33
34

✅ Runs and optimized.  
🟨 Runs and correct but not optimized to green yet.  
❌ Does not pass accuracy test or does not run.