Instructions to use trl-internal-testing/tiny-Olmo3ForCausalLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use trl-internal-testing/tiny-Olmo3ForCausalLM with Transformers:

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

pipe = pipeline("text-generation", model="trl-internal-testing/tiny-Olmo3ForCausalLM")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("trl-internal-testing/tiny-Olmo3ForCausalLM")
model = AutoModelForMultimodalLM.from_pretrained("trl-internal-testing/tiny-Olmo3ForCausalLM")
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-Olmo3ForCausalLM 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-Olmo3ForCausalLM"
# 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-Olmo3ForCausalLM",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/trl-internal-testing/tiny-Olmo3ForCausalLM

SGLang

How to use trl-internal-testing/tiny-Olmo3ForCausalLM 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-Olmo3ForCausalLM" \
    --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-Olmo3ForCausalLM",
		"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-Olmo3ForCausalLM" \
        --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-Olmo3ForCausalLM",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use trl-internal-testing/tiny-Olmo3ForCausalLM with Docker Model Runner:
```
docker model run hf.co/trl-internal-testing/tiny-Olmo3ForCausalLM
```

Upload Olmo3ForCausalLM

by qgallouedec HF Staff - opened 4 days ago

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+15

-8

Files changed (2) hide show

config.json +14 -7
model.safetensors +1 -1

config.json CHANGED Viewed

@@ -5,7 +5,7 @@
   "attention_bias": false,
   "attention_dropout": 0.0,
   "dtype": "bfloat16",
-  "eos_token_id": 50279,
   "hidden_act": "silu",
   "hidden_size": 8,
   "initializer_range": 0.02,
@@ -14,18 +14,25 @@
     "sliding_attention",
     "sliding_attention"
   ],
-  "max_position_embeddings": 2048,
   "model_type": "olmo3",
   "num_attention_heads": 4,
   "num_hidden_layers": 2,
   "num_key_value_heads": 2,
-  "pad_token_id": 1,
-  "rms_norm_eps": 1e-05,
-  "rope_scaling": null,
-  "rope_theta": 10000.0,
   "sliding_window": 4096,
   "tie_word_embeddings": false,
   "transformers_version": "4.57.1",
-  "use_cache": true,
   "vocab_size": 100278
 }

   "attention_bias": false,
   "attention_dropout": 0.0,
   "dtype": "bfloat16",
+  "eos_token_id": 100257,
   "hidden_act": "silu",
   "hidden_size": 8,
   "initializer_range": 0.02,
     "sliding_attention",
     "sliding_attention"
   ],
+  "max_position_embeddings": 65536,
   "model_type": "olmo3",
   "num_attention_heads": 4,
   "num_hidden_layers": 2,
   "num_key_value_heads": 2,
+  "pad_token_id": 100277,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "attention_factor": 1.2079441541679836,
+    "beta_fast": 32.0,
+    "beta_slow": 1.0,
+    "factor": 8.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "yarn"
+  },
+  "rope_theta": 500000,
   "sliding_window": 4096,
   "tie_word_embeddings": false,
   "transformers_version": "4.57.1",
+  "use_cache": false,
   "vocab_size": 100278
 }

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:856489049a6e4e349febe48df59a7458c7e1eba63c7303ccd62ed0074b604530
 size 3215480

 version https://git-lfs.github.com/spec/v1
+oid sha256:870962c27e1cf3d0a3f37eea86e4e731cffb473899634e5f565ff337a7b31e29
 size 3215480