failed with status code 500 while running Skyvern locally with Ollama

[The-BlackSmith872]

Hey guys, how you doing?
I'm trying to set up and test Skyvern locally (via Docker) with Ollama's local LLMs.
I pulled the latest Skyvern image.
These are my only modifications to docker-compose.yml (uncommented lines):

  - LLM_KEY=OLLAMA  
  - ENABLE_OLLAMA=true
  - OLLAMA_MODEL=qwen2.5:7b
  - OLLAMA_SERVER_URL=http://host.docker.internal:11434

I also tried ollama/qwen2.5:7b as the OLLAMA_MODEL (didn't work)

And these are the logs:

docker compose logs skyvern
skyvern-1 | 2025-06-22T15:39:03.250014Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:03.250122Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:03.268010Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:03.268101Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | INFO [alembic.runtime.migration] Context impl PostgresqlImpl.
skyvern-1 | INFO [alembic.runtime.migration] Will assume transactional DDL.
skyvern-1 | Alembic mode: online
skyvern-1 | 2025-06-22T15:39:07.380073Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:07.380222Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:07.401273Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:07.401426Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | INFO [alembic.runtime.migration] Context impl PostgresqlImpl.
skyvern-1 | INFO [alembic.runtime.migration] Will assume transactional DDL.
skyvern-1 | Alembic mode: online
skyvern-1 | No new upgrade operations detected.
skyvern-1 | Starting Xvfb...
skyvern-1 | 2025-06-22T15:39:12.537296Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:12.537416Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:12.556946Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:12.557058Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:13.023674Z [info ] Agent server starting. host=0.0.0.0 port=8000
skyvern-1 | INFO: Will watch for changes in these directories: ['/app']
skyvern-1 | INFO: Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
skyvern-1 | INFO: Started reloader process [45] using WatchFiles
skyvern-1 | 2025-06-22T15:39:16.370500Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:16.370630Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:16.391252Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:16.391399Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA

C:\Users\Artin\Desktop\1 work here\Skyvern\skyvern>docker compose logs skyvern
skyvern-1 | 2025-06-22T15:39:03.250014Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:03.250122Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:03.268010Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:03.268101Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | INFO [alembic.runtime.migration] Context impl PostgresqlImpl.
skyvern-1 | INFO [alembic.runtime.migration] Will assume transactional DDL.
skyvern-1 | Alembic mode: online
skyvern-1 | 2025-06-22T15:39:07.380073Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:07.380222Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:07.401273Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:07.401426Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | INFO [alembic.runtime.migration] Context impl PostgresqlImpl.
skyvern-1 | INFO [alembic.runtime.migration] Will assume transactional DDL.
skyvern-1 | Alembic mode: online
skyvern-1 | No new upgrade operations detected.
skyvern-1 | Starting Xvfb...
skyvern-1 | 2025-06-22T15:39:12.537296Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:12.537416Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:12.556946Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:12.557058Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:13.023674Z [info ] Agent server starting. host=0.0.0.0 port=8000
skyvern-1 | INFO: Will watch for changes in these directories: ['/app']
skyvern-1 | INFO: Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
skyvern-1 | INFO: Started reloader process [45] using WatchFiles
skyvern-1 | 2025-06-22T15:39:16.370500Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:16.370630Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:16.391252Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 | 2025-06-22T15:39:16.391399Z [info ] Using general model configuration for unknown LLM key llm_key=OLLAMA
skyvern-1 |
skyvern-1 | Provider List: https://docs.litellm.ai/docs/providers
skyvern-1 |
skyvern-1 | 2025-06-22T15:39:56.179532Z [error ] LLM request failed unexpectedly llm_key=OLLAMA
skyvern-1 | Traceback (most recent call last):
skyvern-1 | File "/app/skyvern/forge/sdk/api/llm/api_handler_factory.py", line 312, in llm_api_handler
skyvern-1 | response = await litellm.acompletion(
skyvern-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/utils.py", line 1452, in wrapper_async
skyvern-1 | raise e
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/utils.py", line 1313, in wrapper_async
skyvern-1 | result = await original_function(*args, **kwargs)
skyvern-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/main.py", line 446, in acompletion
skyvern-1 | _, custom_llm_provider, _, _ = get_llm_provider(
skyvern-1 | ^^^^^^^^^^^^^^^^^
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 358, in get_llm_provider
skyvern-1 | raise e
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 335, in get_llm_provider
skyvern-1 | raise litellm.exceptions.BadRequestError( # type: ignore
skyvern-1 | litellm.exceptions.BadRequestError: litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=OLLAMA
skyvern-1 | Pass model as E.g. For 'Huggingface' inference endpoints pass in completion(model='huggingface/starcoder',..) Learn more: https://docs.litellm.ai/docs/providers
skyvern-1 | 2025-06-22T15:39:56.180439Z [error ] LLM failure to initialize task v2
skyvern-1 | Traceback (most recent call last):
skyvern-1 | File "/app/skyvern/forge/sdk/api/llm/api_handler_factory.py", line 312, in llm_api_handler
skyvern-1 | response = await litellm.acompletion(
skyvern-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/utils.py", line 1452, in wrapper_async
skyvern-1 | raise e
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/utils.py", line 1313, in wrapper_async
skyvern-1 | result = await original_function(*args, **kwargs)
skyvern-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/main.py", line 446, in acompletion
skyvern-1 | _, custom_llm_provider, _, _ = get_llm_provider(
skyvern-1 | ^^^^^^^^^^^^^^^^^
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 358, in get_llm_provider
skyvern-1 | raise e
skyvern-1 | File "/usr/local/lib/python3.11/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 335, in get_llm_provider
skyvern-1 | raise litellm.exceptions.BadRequestError( # type: ignore
skyvern-1 | litellm.exceptions.BadRequestError: litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=OLLAMA
skyvern-1 | Pass model as E.g. For 'Huggingface' inference endpoints pass in completion(model='huggingface/starcoder',..) Learn more: https://docs.litellm.ai/docs/providers
skyvern-1 |
skyvern-1 | The above exception was the direct cause of the following exception:
skyvern-1 |
skyvern-1 | Traceback (most recent call last):
skyvern-1 | File "/app/skyvern/forge/sdk/routes/agent_protocol.py", line 1341, in run_task_v2
skyvern-1 | task_v2 = await task_v2_service.initialize_task_v2(
skyvern-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
skyvern-1 | File "/app/skyvern/services/task_v2_service.py", line 125, in initialize_task_v2
skyvern-1 | metadata_response = await app.LLM_API_HANDLER(
skyvern-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^
skyvern-1 | File "/app/skyvern/forge/sdk/api/llm/api_handler_factory.py", line 338, in llm_api_handler
skyvern-1 | raise LLMProviderError(llm_key) from e
skyvern-1 | skyvern.forge.sdk.api.llm.exceptions.LLMProviderError: Error while using LLMProvider OLLAMA

---

Would really appreciate your help on this, thank you in advance.