Instructions to use microsoft/FrogBoss-32B-2510 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/FrogBoss-32B-2510 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="microsoft/FrogBoss-32B-2510")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/FrogBoss-32B-2510")
model = AutoModelForCausalLM.from_pretrained("microsoft/FrogBoss-32B-2510")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use microsoft/FrogBoss-32B-2510 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/FrogBoss-32B-2510"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/FrogBoss-32B-2510",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/microsoft/FrogBoss-32B-2510

SGLang

How to use microsoft/FrogBoss-32B-2510 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "microsoft/FrogBoss-32B-2510" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/FrogBoss-32B-2510",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "microsoft/FrogBoss-32B-2510" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/FrogBoss-32B-2510",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use microsoft/FrogBoss-32B-2510 with Docker Model Runner:
```
docker model run hf.co/microsoft/FrogBoss-32B-2510
```

chsingh commited on Jan 22

Commit

cc93095

verified ·

1 Parent(s): 5329af4

Update frogboss_r2egym_parser.py

Browse files

Files changed (1) hide show

frogboss_r2egym_parser.py +4 -4

frogboss_r2egym_parser.py CHANGED Viewed

@@ -4,10 +4,10 @@ Custom tool parser for vLLM with R2E-gym XML format.
 Same as frogboss_default_parser but handles XML format instead of JSON.
 Usage:
-vllm serve microsoft/FrogBoss-2510 \
     --tensor-parallel-size 4 \
         --enable-auto-tool-choice \
-            --tool-parser-plugin frogboss_r2egym_parser.py \
                 --tool-call-parser froggy \
                     --enable-log-requests \
                         --enable-log-outputs \
@@ -28,8 +28,8 @@ from vllm.entrypoints.openai.protocol import (
     FunctionCall,
     ToolCall,
 )
-from vllm.entrypoints.openai.tool_parsers import ToolParser, ToolParserManager
-from vllm.entrypoints.openai.tool_parsers.abstract_tool_parser import (
     ExtractedToolCallInformation,
 )
 from vllm.transformers_utils.tokenizer import AnyTokenizer

 Same as frogboss_default_parser but handles XML format instead of JSON.
 Usage:
+vllm serve microsoft/FrogBoss-32B-2510 \
     --tensor-parallel-size 4 \
         --enable-auto-tool-choice \
+            --tool-parser-plugin ./Froggy-Training/src/vllm/frogboss_r2egym_parser.py \
                 --tool-call-parser froggy \
                     --enable-log-requests \
                         --enable-log-outputs \
     FunctionCall,
     ToolCall,
 )
+from vllm.tool_parsers import ToolParser, ToolParserManager
+from vllm.tool_parsers.abstract_tool_parser import (
     ExtractedToolCallInformation,
 )
 from vllm.transformers_utils.tokenizer import AnyTokenizer