Instructions to use RecursiveSI/flint with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RecursiveSI/flint with Transformers:

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

pipe = pipeline("text-generation", model="RecursiveSI/flint")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RecursiveSI/flint")
model = AutoModelForCausalLM.from_pretrained("RecursiveSI/flint", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RecursiveSI/flint with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/RecursiveSI/flint

SGLang

How to use RecursiveSI/flint 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 "RecursiveSI/flint" \
    --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": "RecursiveSI/flint",
		"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 "RecursiveSI/flint" \
        --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": "RecursiveSI/flint",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RecursiveSI/flint with Docker Model Runner:
```
docker model run hf.co/RecursiveSI/flint
```

flint / sglang /serve_node_8x.sh

lvmopt

Initial release

2a247be 1 day ago

Raw History Blame Contribute Delete

1.08 kB

#!/usr/bin/env bash
# Eight Flint replicas on one 8-GPU node, one GPU each, on ports BASE_PORT..BASE_PORT+7.
# If one replica exits, all are stopped. Put a router in front of all replicas across nodes (on a
# port outside those ranges), e.g. for 2 nodes:
#
#   python3 -m sglang_router.launch_router --host 0.0.0.0 --port 8000 --policy cache_aware \
#     --worker-urls http://node0:30000 ... http://node0:30007 http://node1:30000 ... http://node1:30007
#
#   MODEL=/path/to/flint BASE_PORT=30000 bash serve_node_8x.sh            # Blackwell
#   MODEL=/path/to/flint SERVE_SCRIPT=serve_h200.sh bash serve_node_8x.sh  # H200
#
# Extra arguments are passed through to every replica's sglang.launch_server.
set -euo pipefail

BASE_PORT="${BASE_PORT:-30000}"
SERVE_SCRIPT="${SERVE_SCRIPT:-serve_b300.sh}"
here="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"

pids=()
for gpu in 0 1 2 3 4 5 6 7; do
  CUDA_VISIBLE_DEVICES="$gpu" PORT="$((BASE_PORT + gpu))" bash "$here/$SERVE_SCRIPT" "$@" &
  pids+=("$!")
done
trap 'kill "${pids[@]}" 2>/dev/null; wait' EXIT INT TERM
wait -n "${pids[@]}"

	#!/usr/bin/env bash
	# Eight Flint replicas on one 8-GPU node, one GPU each, on ports BASE_PORT..BASE_PORT+7.
	# If one replica exits, all are stopped. Put a router in front of all replicas across nodes (on a
	# port outside those ranges), e.g. for 2 nodes:
	#
	# python3 -m sglang_router.launch_router --host 0.0.0.0 --port 8000 --policy cache_aware \
	# --worker-urls http://node0:30000 ... http://node0:30007 http://node1:30000 ... http://node1:30007
	#
	# MODEL=/path/to/flint BASE_PORT=30000 bash serve_node_8x.sh # Blackwell
	# MODEL=/path/to/flint SERVE_SCRIPT=serve_h200.sh bash serve_node_8x.sh # H200
	#
	# Extra arguments are passed through to every replica's sglang.launch_server.
	set -euo pipefail

	BASE_PORT="${BASE_PORT:-30000}"
	SERVE_SCRIPT="${SERVE_SCRIPT:-serve_b300.sh}"
	here="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"

	pids=()
	for gpu in 0 1 2 3 4 5 6 7; do
	CUDA_VISIBLE_DEVICES="$gpu" PORT="$((BASE_PORT + gpu))" bash "$here/$SERVE_SCRIPT" "$@" &
	pids+=("$!")
	done
	trap 'kill "${pids[@]}" 2>/dev/null; wait' EXIT INT TERM
	wait -n "${pids[@]}"