Instructions to use aisquared/bolt-instruct-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use aisquared/bolt-instruct-v2 with Transformers:

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

pipe = pipeline("image-text-to-text", model="aisquared/bolt-instruct-v2")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("aisquared/bolt-instruct-v2")
model = AutoModelForMultimodalLM.from_pretrained("aisquared/bolt-instruct-v2", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use aisquared/bolt-instruct-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aisquared/bolt-instruct-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aisquared/bolt-instruct-v2",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/aisquared/bolt-instruct-v2

SGLang

How to use aisquared/bolt-instruct-v2 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 "aisquared/bolt-instruct-v2" \
    --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": "aisquared/bolt-instruct-v2",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "aisquared/bolt-instruct-v2" \
        --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": "aisquared/bolt-instruct-v2",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use aisquared/bolt-instruct-v2 with Docker Model Runner:
```
docker model run hf.co/aisquared/bolt-instruct-v2
```

Update README.md

by iansotnek - opened 5 days ago

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+12

-4

Files changed (1) hide show

README.md +12 -4

README.md CHANGED Viewed

@@ -62,6 +62,14 @@ Key design goals:
 Bolt Instruct V2 is a LoRA supervised fine-tune, trained to combine concise reasoning with precise tool use. The goal was to reduce unnecessary deliberation while preserving the ability to make tool calls. We chose supervised fine-tuning because we could demonstrate the behaviour we wanted directly: brief reasoning followed by the appropriate tool calls, including grouped calls for parallel execution.
 The training corpus contains 7,646 multi-step agentic sessions, filtered to include tool schemas, at least one reasoning block, and a genuine user request. Reasoning before tool calls had a median length of 29 tokens, and 91% of tool calls were preceded by reasoning. The corpus also included 14,204 turns with multiple tool calls, representing 17.4% of tool-calling turns and occurring in 47% of sessions. Sessions contained a median of 13 tool calls, with 63% containing at least 10. Together, these examples demonstrated concise reasoning, parallel tool calling, and sustained multi-step tool use.
 ---
 # Intended Use
@@ -109,13 +117,15 @@ print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))
 We evaluated Bolt Instruct V2 on raw performance on general-purpose and agentic tasks, as well as reasoning efficiency and time to complete a given task.
 ### Evaluation results for aisquared/bolt-instruct-v2:
 Measured with the sampling configuration published for Muse Glimmer
 **temperature 1.0 / top_p 0.95 / top_k 64**, high reasoning strength
 served with vLLM (`--generation-config auto`).
 ### Efficiency — per task
 | Metric | Benchmark | Muse Glimmer 30B(High) | **Bolt Instruct V2** | Gemma 4 31B (think on) |
@@ -128,7 +138,6 @@ served with vLLM (`--generation-config auto`).
 | | AutomationBench | 9,910 | **5,988** | 14,839 |
 | **Latency** | BFCL (mean) | 19.76 s | **12.60 s** | 7.47 s |
 | | τ²-bench | 64.9 s | **51.4 s** | 129.8 s |
-| | AutomationBench | 371.4 s | **342.1 s** | 1,079.8 s |
 ### Agentic — multi-turn
@@ -136,7 +145,6 @@ served with vLLM (`--generation-config auto`).
 |---|---|---|---|
 | τ²-bench @ temperature 0 | 85.37 | 76.83 | 84.15 |
 | τ²-bench @ published sampling ¹ | 34.45 | 30.64 | 85.21 |
-| AutomationBench ² | 45.38 | 27.70 | 48.89 |
 ### Agentic — single-turn tool calling
@@ -159,7 +167,7 @@ served with vLLM (`--generation-config auto`).
 Reasoning-token reduction versus Muse Glimmer 30B: **49.8%** (BFCL),
-**42.2%** (τ²-bench), **68.2%** (AutomationBench).
 **Reproduction.** τ²-bench and AutomationBench under their official

 Bolt Instruct V2 is a LoRA supervised fine-tune, trained to combine concise reasoning with precise tool use. The goal was to reduce unnecessary deliberation while preserving the ability to make tool calls. We chose supervised fine-tuning because we could demonstrate the behaviour we wanted directly: brief reasoning followed by the appropriate tool calls, including grouped calls for parallel execution.
 The training corpus contains 7,646 multi-step agentic sessions, filtered to include tool schemas, at least one reasoning block, and a genuine user request. Reasoning before tool calls had a median length of 29 tokens, and 91% of tool calls were preceded by reasoning. The corpus also included 14,204 turns with multiple tool calls, representing 17.4% of tool-calling turns and occurring in 47% of sessions. Sessions contained a median of 13 tool calls, with 63% containing at least 10. Together, these examples demonstrated concise reasoning, parallel tool calling, and sustained multi-step tool use.
+### Training Data
+bolt-instruct-v2 was trained on a collected and generated set of data without commercial or other usage restrictions.
+### Training Hardware
+bolt-instruct-v2 was trained for 18 hours on a 2xH200 cluster.
 ---
 # Intended Use
 We evaluated Bolt Instruct V2 on raw performance on general-purpose and agentic tasks, as well as reasoning efficiency and time to complete a given task.
 ### Evaluation results for aisquared/bolt-instruct-v2:
 Measured with the sampling configuration published for Muse Glimmer
 **temperature 1.0 / top_p 0.95 / top_k 64**, high reasoning strength
 served with vLLM (`--generation-config auto`).
+![efficiency_donuts](https://cdn-uploads.huggingface.co/production/uploads/6271334067ff0a433db23501/-U2ok6mZQ8si_7Ncx7gFz.png)
 ### Efficiency — per task
 | Metric | Benchmark | Muse Glimmer 30B(High) | **Bolt Instruct V2** | Gemma 4 31B (think on) |
 | | AutomationBench | 9,910 | **5,988** | 14,839 |
 | **Latency** | BFCL (mean) | 19.76 s | **12.60 s** | 7.47 s |
 | | τ²-bench | 64.9 s | **51.4 s** | 129.8 s |
 ### Agentic — multi-turn
 |---|---|---|---|
 | τ²-bench @ temperature 0 | 85.37 | 76.83 | 84.15 |
 | τ²-bench @ published sampling ¹ | 34.45 | 30.64 | 85.21 |
 ### Agentic — single-turn tool calling
 Reasoning-token reduction versus Muse Glimmer 30B: **49.8%** (BFCL),
+**42.2%** (τ²-bench).
 **Reproduction.** τ²-bench and AutomationBench under their official