A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 12, 23:30 UTC
Add CPU support model (#28697)
Signed-off-by: Tsai, Louie <louie.tsai@intel.com>
Louie Tsai committed
Nov 18, 2025 at 23:47 UTC
ae4821a1086325decbc801d3292dee42e42549bb
1 file changed
+26
docs/models/hardware_supported_models/cpu.md
new
+26
@@ -0,0 +1,26 @@
1
+# CPU - Intel® Xeon®
2
+
3
+## Supported Models
4
+
5
+### Text-only Language Models
6
+
7
+| Model | Architecture | Supported |
8
+|--------------------------------------|-------------------------------------------|-----------|
9
+| meta-llama/Llama-3.1 / 3.3 | LlamaForCausalLM | ✅ |
10
+| meta-llama/Llama-4-Scout | Llama4ForConditionalGeneration | ✅ |
11
+| meta-llama/Llama-4-Maverick | Llama4ForConditionalGeneration | ✅ |
12
+| ibm-granite/granite (Granite-MOE) | GraniteMoeForCausalLM | ✅ |
13
+| Qwen/Qwen3 | Qwen3ForCausalLM | ✅ |
14
+| zai-org/GLM-4.5 | GLMForCausalLM | ✅ |
15
+| google/gemma | GemmaForCausalLM | ✅ |
16
+
17
+### Multimodal Language Models
18
+
19
+| Model | Architecture | Supported |
20
+|--------------------------------------|-------------------------------------------|-----------|
21
+| Qwen/Qwen2.5-VL | Qwen2VLForConditionalGeneration | ✅ |
22
+| openai/whisper | WhisperForConditionalGeneration | ✅ |
23
+
24
+✅ Runs and optimized.
25
+🟨 Runs and correct but not optimized to green yet.
26
+❌ Does not pass accuracy test or does not run.