%%capture

!pip install -q litellm  python-dotenv

import os, sys

## Recommended way of getting configuration
if os.getenv("COLAB_RELEASE_TAG"):
   print("Running in Colab")
   from google.colab import userdata
   NEBIUS_API_KEY = userdata.get('NEBIUS_API_KEY')
else:
   print("NOT running in Colab")
   from dotenv import load_dotenv
   this_dir = os.path.abspath('')
   parent_dir = os.path.dirname(this_dir)
   sys.path.append (os.path.abspath (parent_dir))
   load_dotenv()
   NEBIUS_API_KEY = os.getenv('NEBIUS_API_KEY')

## quick hack (not recommended) - you can hardcode the config key here
# NEBIUS_API_KEY = "your_key_here"

if NEBIUS_API_KEY:
  print ('✅ NEBIUS_API_KEY found')
  os.environ['NEBIUS_API_KEY'] = NEBIUS_API_KEY
else:
  raise RuntimeError ('❌ NEBIUS_API_KEY NOT found')

NOT running in Colab
✅ NEBIUS_API_KEY found

%%time

import os
from litellm import completion

response = completion(
    model="nebius/openai/gpt-oss-20b-Instruct-2507",
    messages=[
        {
            "role": "user",
            "content": "What is the capital of Netherlands?",
        }
    ],
    temperature=0.1,  # either set temperature or `top_p`
)

print ('----model answer -----')
print (response.choices[0].message.content)
print ('\n----- full response ----')
print(response.to_json())
print ('---------')

----model answer -----
The capital of the Netherlands is Amsterdam.

----- full response ----
{
  "id": "chatcmpl-63c63b527d454383946b3481709c2b81",
  "created": 1762232294,
  "model": "nebius/openai/gpt-oss-20b-Instruct-2507",
  "object": "chat.completion",
  "system_fingerprint": null,
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": "The capital of the Netherlands is Amsterdam.",
        "role": "assistant",
        "tool_calls": null,
        "function_call": null
      },
      "provider_specific_fields": {
        "stop_reason": null,
        "token_ids": null
      }
    }
  ],
  "usage": {
    "completion_tokens": 9,
    "prompt_tokens": 15,
    "total_tokens": 24,
    "completion_tokens_details": null,
    "prompt_tokens_details": null
  },
  "service_tier": null,
  "prompt_logprobs": null,
  "prompt_token_ids": null,
  "kv_transfer_params": null
}
---------
CPU times: user 820 ms, sys: 194 ms, total: 1.01 s
Wall time: 4.25 s

Run models on Nebius Token Factory (via LiteLLM API)¶

References and Acknowledgements¶

Pre requisites¶

1 - Install Dependencies¶

2 - Load Configuration¶

3 - Pick a Model¶

4 - Run the Model¶

5 - Try Your Queries¶