Instructions to use trl-internal-testing/tiny-FalconMambaForCausalLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use trl-internal-testing/tiny-FalconMambaForCausalLM with Transformers:

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

pipe = pipeline("text-generation", model="trl-internal-testing/tiny-FalconMambaForCausalLM")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("trl-internal-testing/tiny-FalconMambaForCausalLM")
model = AutoModelForCausalLM.from_pretrained("trl-internal-testing/tiny-FalconMambaForCausalLM", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use trl-internal-testing/tiny-FalconMambaForCausalLM with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/trl-internal-testing/tiny-FalconMambaForCausalLM

SGLang

How to use trl-internal-testing/tiny-FalconMambaForCausalLM 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 "trl-internal-testing/tiny-FalconMambaForCausalLM" \
    --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": "trl-internal-testing/tiny-FalconMambaForCausalLM",
		"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 "trl-internal-testing/tiny-FalconMambaForCausalLM" \
        --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": "trl-internal-testing/tiny-FalconMambaForCausalLM",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use trl-internal-testing/tiny-FalconMambaForCausalLM with Docker Model Runner:
```
docker model run hf.co/trl-internal-testing/tiny-FalconMambaForCausalLM
```

Upload FalconMambaForCausalLM

by albertvillanova HF Staff - opened about 9 hours ago

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

-6

Files changed (2) hide show

config.json +3 -4
model.safetensors +2 -2

config.json CHANGED Viewed

@@ -2,10 +2,10 @@
   "architectures": [
     "FalconMambaForCausalLM"
   ],
-  "bos_token_id": 0,
   "conv_kernel": 4,
   "dtype": "bfloat16",
-  "eos_token_id": 0,
   "expand": 2,
   "hidden_act": "silu",
   "hidden_size": 8,
@@ -14,13 +14,12 @@
   "layer_norm_epsilon": 1e-05,
   "mixer_rms_eps": 1e-06,
   "model_type": "falcon_mamba",
-  "num_attention_heads": 4,
   "num_hidden_layers": 2,
-  "num_key_value_heads": 2,
   "pad_token_id": 0,
   "rescale_prenorm_residual": false,
   "residual_in_fp32": true,
   "state_size": 16,
   "time_step_floor": 0.0001,
   "time_step_init_scheme": "random",
   "time_step_max": 0.1,

   "architectures": [
     "FalconMambaForCausalLM"
   ],
+  "bos_token_id": 8,
   "conv_kernel": 4,
   "dtype": "bfloat16",
+  "eos_token_id": 11,
   "expand": 2,
   "hidden_act": "silu",
   "hidden_size": 8,
   "layer_norm_epsilon": 1e-05,
   "mixer_rms_eps": 1e-06,
   "model_type": "falcon_mamba",
   "num_hidden_layers": 2,
   "pad_token_id": 0,
   "rescale_prenorm_residual": false,
   "residual_in_fp32": true,
   "state_size": 16,
+  "tie_word_embeddings": false,
   "time_step_floor": 0.0001,
   "time_step_init_scheme": "random",
   "time_step_max": 0.1,

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:058b95590b4d09739a25ffa74f2b4728ce7b4f2b6e108903751774a26ebc88e9
-size 1053056

 version https://git-lfs.github.com/spec/v1
+oid sha256:03ca9f4c1b20fd93e68b997c712ef8d8b82b68b7901ca4678615f8a086824f44
+size 2093528