chore: import upstream snapshot with attribution

examples/infer/transformers/batch_ddp.sh

@@ -0,0 +1,9 @@
+# 18GB
+NPROC_PER_NODE=4 \
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+swift infer \
+    --model Qwen/Qwen2.5-1.5B-Instruct \
+    --infer_backend transformers \
+    --val_dataset AI-ModelScope/alpaca-gpt4-data-zh#1000 \
+    --max_batch_size 16 \
+    --max_new_tokens 512

examples/infer/transformers/bert.sh

@@ -0,0 +1,8 @@
+# Since `swift/test_lora` is trained by swift and contains an `args.json` file,
+# there is no need to explicitly set `--model`, `--system`, etc., as they will be automatically read.
+# To disable this behavior, please set `--load_args false`.
+CUDA_VISIBLE_DEVICES=0 \
+swift infer \
+    --adapters swift/test_bert \
+    --truncation_strategy right \
+    --max_length 512

examples/infer/transformers/lora.sh

@@ -0,0 +1,10 @@
+# Since `swift/test_lora` is trained by swift and contains an `args.json` file,
+# there is no need to explicitly set `--model`, `--system`, etc., as they will be automatically read.
+# To disable this behavior, please set `--load_args false`.
+CUDA_VISIBLE_DEVICES=0 \
+swift infer \
+    --adapters swift/test_lora \
+    --infer_backend transformers \
+    --stream true \
+    --temperature 0 \
+    --max_new_tokens 2048

examples/infer/transformers/mllm_device_map.sh

@@ -0,0 +1,9 @@
+NPROC_PER_NODE=2 \
+CUDA_VISIBLE_DEVICES=0,1,2,3 \
+MAX_PIXELS=1003520 \
+swift infer \
+    --model Qwen/Qwen2.5-VL-3B-Instruct \
+    --infer_backend transformers \
+    --val_dataset AI-ModelScope/LaTeX_OCR#1000 \
+    --max_batch_size 16 \
+    --max_new_tokens 512

examples/infer/transformers/prm.sh

@@ -0,0 +1,4 @@
+CUDA_VISIBLE_DEVICES=0 \
+swift infer \
+    --model Qwen/Qwen2.5-Math-PRM-7B \
+    --infer_backend transformers

examples/infer/transformers/reward_model.sh

@@ -0,0 +1,5 @@
+CUDA_VISIBLE_DEVICES=0 \
+swift infer \
+    --model Shanghai_AI_Laboratory/internlm2-1_8b-reward \
+    --val_dataset AI-ModelScope/alpaca-gpt4-data-zh#1000 \
+    --max_batch_size 64