[train] KTransformers SFT as backend engine for LLaMA-Factory (#9400)

Co-authored-by: jimmy128 <jimmy128@noreply.gitcode.com>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>

examples/inference/deepseek2_lora_sft_kt.yaml

@@ -0,0 +1,10 @@
+model_name_or_path: deepseek-ai/DeepSeek-V2-Lite
+adapter_name_or_path: saves/Kllama_deepseekV2
+template: deepseek
+infer_backend: ktransformers  # choices: [huggingface, vllm, sglang, ktransformers]
+trust_remote_code: true
+
+use_kt: true # use KTransformers as LoRA sft backend to inference
+kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V2-Lite-Chat-sft-amx.yaml
+cpu_infer: 32
+chunk_size: 8192

examples/inference/deepseek3_kt.yaml

@@ -0,0 +1,9 @@
+model_name_or_path: opensourcerelease/DeepSeek-V3-bf16
+template: deepseek
+infer_backend: ktransformers  # choices: [huggingface, vllm, sglang, ktransformers]
+trust_remote_code: true
+
+use_kt: true # use KTransformers as LoRA sft backend to inference
+kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V3-Chat-sft-amx-multi-gpu.yaml
+cpu_infer: 32
+chunk_size: 8192

examples/inference/deepseek3_lora_sft_kt.yaml

@@ -0,0 +1,10 @@
+model_name_or_path: opensourcerelease/DeepSeek-V3-bf16
+adapter_name_or_path: saves/Kllama_deepseekV3
+template: deepseek
+infer_backend: ktransformers  # choices: [huggingface, vllm, sglang, ktransformers]
+trust_remote_code: true
+
+use_kt: true # use KTransformers as LoRA sft backend to inference
+kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V3-Chat-sft-amx-multi-gpu.yaml
+cpu_infer: 32
+chunk_size: 8192

examples/inference/llama3.yaml

@@ -1,4 +1,4 @@
 model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct
 template: llama3
-infer_backend: huggingface  # choices: [huggingface, vllm, sglang]
+infer_backend: huggingface  # choices: [huggingface, vllm, sglang, ktransformers]
 trust_remote_code: true

examples/inference/llama3_full_sft.yaml

@@ -1,4 +1,4 @@
 model_name_or_path: saves/llama3-8b/full/sft
 template: llama3
-infer_backend: huggingface  # choices: [huggingface, vllm, sglang]
+infer_backend: huggingface  # choices: [huggingface, vllm, sglang, ktransformers]
 trust_remote_code: true

examples/inference/llama3_lora_sft.yaml

@@ -1,5 +1,5 @@
 model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct
 adapter_name_or_path: saves/llama3-8b/lora/sft
 template: llama3
-infer_backend: huggingface  # choices: [huggingface, vllm, sglang]
+infer_backend: huggingface  # choices: [huggingface, vllm, sglang, ktransformers]
 trust_remote_code: true

examples/inference/qwen2_5vl.yaml

@@ -1,4 +1,4 @@
 model_name_or_path: Qwen/Qwen2.5-VL-7B-Instruct
 template: qwen2_vl
-infer_backend: huggingface  # choices: [huggingface, vllm, sglang]
+infer_backend: huggingface  # choices: [huggingface, vllm, sglang, ktransformers]
 trust_remote_code: true