Ollama embedding component does not show the model listed

[pulkit21aug]

Bug Description

Ollama embedding component does not show the model listed

Reproduction

Ollama embedding component does not show the model listed

curl http://localhost:11434/
Ollama is running

curl http://localhost:11434/api/tags
{
"models": [
{
"name": "gemma3:4b",
"model": "gemma3:4b",
"modified_at": "2025-12-21T15:43:44.3658294+05:30",
"size": 3338801804,
"digest": "a2af6cc3eb7fa8be8504abaf9b04e88f17a119ec3f04a3addf55f92841195f5a",
"details": {
"parent_model": "",
"format": "gguf",
"family": "gemma3",
"families": [
"gemma3"
],
"parameter_size": "4.3B",
"quantization_level": "Q4_K_M"
}
},
{
"name": "deepseek-r1:8b",
"model": "deepseek-r1:8b",
"modified_at": "2025-12-21T11:44:46.4384762+05:30",
"size": 5225376047,
"digest": "6995872bfe4c521a67b32da386cd21d5c6e819b6e0d62f79f64ec83be99f5763",
"details": {
"parent_model": "",
"format": "gguf",
"family": "qwen3",
"families": [
"qwen3"
],
"parameter_size": "8.2B",
"quantization_level": "Q4_K_M"
}
}
]
}

There are 2 models installed however embedding does not show them. Rfreshed multiple times but still not working

Expected behavior

Models should come in dropdown

Who can help?

No response

Operating System

windows

Langflow Version

latest

Python Version

None

Screenshot

No response

Flow File

No response

[coderabbitai]

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

• Create Plan

Examples

• Example 1
• Example 2

---

🔗 Similar Issues

Possible Duplicates

• 10331

Related Issues

• Cannot Run ollama embeddings #10331
• Ollama doesn't show models #8172
• Ollama generation component fails to load models, while Embeddings component works fine #9488
• Ollama Embedding Component doesn't retain the base URL #8881

👤 Suggested Assignees

• avi2000
• gregers666
• rojosan0617
• factor3

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord or schedule a call!

[Olaiwonismail]

Are you working on this
if not can i work on it?

[MugheesMehdi07]

I'm working on this issue. I've identified the problem and created a fix.

Problem: The Ollama embeddings component was filtering models too strictly, only showing models that matched hardcoded embedding model names from OLLAMA_EMBEDDING_MODELS. This prevented custom models like gemma3:4b and deepseek-r1:8b from appearing in the dropdown.

Solution: Removed the strict filtering logic so all available models from the Ollama API are displayed. The Ollama API will handle validation when the model is actually used.

Changes:

• Removed strict filtering in get_model() method
• Added comprehensive tests to verify all models are returned
• Tests specifically verify that custom models (gemma3:4b, deepseek-r1:8b) are included

PR will be created shortly. Please assign

[andifilhohub]

Hi, @MugheesMehdi07 @pulkit21aug @Olaiwonismail

Could you please try deactivating the "Tool mode enable" on the component and then attempt to load the list?

[pulkit21aug]

@andifilhohub tool mode option is not visibile for embedding component of ollama bundle

[Empreiteiro]

@pulkit21aug Currently, the Embedding component is only prepared to list specific embedding models, and does not support multimodal models.

nomic-embed-text:latest
mxbai-embed-large:latest
all-MiniLM-L6-v
...

[pulkit21aug]

Not sure what is wrong . below is the list of models deployed at my PC

But these models are not getting listed in langflow components - check below

[pulkit21aug]

Finally this one got loaded

[pulkit21aug]

can you fix it that all models get listed instead of developers like me getting confused