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

Libraries

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

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

pipe = pipeline("text-generation", model="trl-internal-testing/tiny-DbrxForCausalLM")
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-DbrxForCausalLM")
model = AutoModelForCausalLM.from_pretrained("trl-internal-testing/tiny-DbrxForCausalLM")
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-DbrxForCausalLM 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-DbrxForCausalLM"
# 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-DbrxForCausalLM",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

SGLang

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

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

Upload DbrxForCausalLM

by albertvillanova HF Staff - opened Oct 6, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+17

-18

Files changed (3) hide show

config.json +14 -15
generation_config.json +1 -1
model.safetensors +2 -2

config.json CHANGED Viewed

@@ -4,38 +4,37 @@
   ],
   "attn_config": {
     "attn_pdrop": 0.0,
-    "clip_qkv": 8,
-    "kv_n_heads": 8,
     "model_type": "",
-    "rope_theta": 500000
   },
-  "d_model": 32,
   "dtype": "float32",
   "emb_pdrop": 0.0,
   "ffn_config": {
     "ffn_act_fn": {
       "name": "silu"
     },
-    "ffn_hidden_size": 10752,
-    "hidden_size": 32,
     "model_type": "",
-    "moe_jitter_eps": 0,
-    "moe_loss_weight": 0.05,
     "moe_normalize_expert_weights": 1.0,
-    "moe_num_experts": 16,
-    "moe_top_k": 4
   },
   "initializer_range": 0.02,
-  "max_seq_len": 32768,
   "model_type": "dbrx",
   "n_heads": 4,
   "n_layers": 2,
-  "num_key_value_heads": 8,
   "output_router_logits": false,
   "resid_pdrop": 0.0,
-  "rope_theta": 10000.0,
   "tie_word_embeddings": false,
-  "transformers_version": "4.57.0.dev0",
   "use_cache": true,
-  "vocab_size": 100352
 }

   ],
   "attn_config": {
     "attn_pdrop": 0.0,
+    "clip_qkv": null,
+    "kv_n_heads": 1,
     "model_type": "",
+    "rope_theta": 10000.0
   },
+  "d_model": 8,
   "dtype": "float32",
   "emb_pdrop": 0.0,
   "ffn_config": {
     "ffn_act_fn": {
       "name": "silu"
     },
+    "ffn_hidden_size": 3584,
     "model_type": "",
+    "moe_jitter_eps": null,
+    "moe_loss_weight": 0.01,
     "moe_normalize_expert_weights": 1.0,
+    "moe_num_experts": 4,
+    "moe_top_k": 1
   },
   "initializer_range": 0.02,
+  "intermediate_size": 32,
+  "max_seq_len": 2048,
   "model_type": "dbrx",
   "n_heads": 4,
   "n_layers": 2,
+  "num_key_value_heads": 2,
   "output_router_logits": false,
   "resid_pdrop": 0.0,
   "tie_word_embeddings": false,
+  "transformers_version": "4.57.0",
   "use_cache": true,
+  "vocab_size": 100280
 }

generation_config.json CHANGED Viewed

@@ -1,4 +1,4 @@
 {
   "_from_model_config": true,
-  "transformers_version": "4.57.0.dev0"
 }

 {
   "_from_model_config": true,
+  "transformers_version": "4.57.0"
 }

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca68fe84a06c2af3f0b51124a49048055af1b5d503cc249915f8f29eca48d1f6
-size 159238944

 version https://git-lfs.github.com/spec/v1
+oid sha256:1d9db7399eeb0f7ebfbf1b716f456a57dfbd914e366c15dc8d230f28e0c636d5
+size 9174240