github-starred/ollama-ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-27 07:57:18 -05:00
Code Issues 2.2k Packages Projects Releases 100 Wiki Activity

hf.co/LiquidAI/LFM2-8B-A1B-GGUF:latest #8334

New Issue
Open
opened 2025-11-12 14:38:24 -06:00 by GiteaMirror · 7 comments
GiteaMirror commented 2025-11-12 14:38:24 -06:00
Owner
Copy Link

Originally created by @pchankh on GitHub (Oct 9, 2025).

We ran into the problem after downloading the model image.
ollama run hf.co/LiquidAI/LFM2-8B-A1B-GGUF:latest
Error: 500 Internal Server Error: unable to load model: /usr/share/ollama/.ollama/models/blobs/sha256-d2185b22630fc68043dac7182f12e86e5ad14990229a90b6c9ad3f4421ddaf82

Is the current model supported yet?

Thanks
Dr Patrick

Originally created by @pchankh on GitHub (Oct 9, 2025). We ran into the problem after downloading the model image. ollama run hf.co/LiquidAI/LFM2-8B-A1B-GGUF:latest Error: 500 Internal Server Error: unable to load model: /usr/share/ollama/.ollama/models/blobs/sha256-d2185b22630fc68043dac7182f12e86e5ad14990229a90b6c9ad3f4421ddaf82 Is the current model supported yet? Thanks Dr Patrick
GiteaMirror added the model label 2025-11-12 14:38:24 -06:00
GiteaMirror commented 2025-11-12 14:38:25 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 9, 2025):

Not yet. Support has been merged into llama.cpp (https://github.com/ggml-org/llama.cpp/pull/16464) so ollama will support it at the next vendor sync.

@rick-github commented on GitHub (Oct 9, 2025): Not yet. Support has been merged into llama.cpp (https://github.com/ggml-org/llama.cpp/pull/16464) so ollama will support it at the next vendor sync.
GiteaMirror commented 2025-11-12 14:38:25 -06:00
Author
Owner
Copy Link

@logxdx commented on GitHub (Oct 14, 2025):

It still doesn't work in 0.12.5

@logxdx commented on GitHub (Oct 14, 2025): It still doesn't work in 0.12.5
GiteaMirror commented 2025-11-12 14:38:25 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 14, 2025):

next vendor sync

@rick-github commented on GitHub (Oct 14, 2025): > next vendor sync
GiteaMirror commented 2025-11-12 14:38:26 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 19, 2025):

Vendor sync done (https://github.com/ollama/ollama/pull/12552) done, the model is supported in ollama 0.12.6.

$ ollama run hf.co/LiquidAI/LFM2-8B-A1B-GGUF:latest
>>> hello
Hello! How can I assist you today? 😊

>>> why is the sky blue?
The sky appears blue due to a phenomenon called **Rayleigh scattering**. Here's how it works:

Sunlight, or white light from the Sun, is made up of many colors (wavelengths), including red, orange, yellow, green, blue, indigo, and violet. When sunlight enters Earth’s atmosphere, it interacts with
air molecules and tiny particles.

- **Shorter wavelengths** like blue and violet are scattered in all directions by the gases and tiny particles in the atmosphere much more effectively than longer wavelengths like red or orange.
- Although violet light is scattered slightly more than blue, our eyes are more sensitive to blue, and some violet light is absorbed by the upper atmosphere. As a result, we predominantly perceive the sky
as **blue** during the day.

At sunrise or sunset, the sky turns red or orange because sunlight passes through more of the atmosphere, scattering away most of the blue and leaving longer wavelengths like red and orange to dominate. 🌅💙


Let me know if you'd like a deeper explanation! 😊

The template is basic and doesn't support tool calling.

@rick-github commented on GitHub (Oct 19, 2025): Vendor sync done (https://github.com/ollama/ollama/pull/12552) done, the model is supported in [ollama 0.12.6](https://github.com/ollama/ollama/releases/tag/v0.12.6). ```console $ ollama run hf.co/LiquidAI/LFM2-8B-A1B-GGUF:latest >>> hello Hello! How can I assist you today? 😊 >>> why is the sky blue? The sky appears blue due to a phenomenon called **Rayleigh scattering**. Here's how it works: Sunlight, or white light from the Sun, is made up of many colors (wavelengths), including red, orange, yellow, green, blue, indigo, and violet. When sunlight enters Earth’s atmosphere, it interacts with air molecules and tiny particles. - **Shorter wavelengths** like blue and violet are scattered in all directions by the gases and tiny particles in the atmosphere much more effectively than longer wavelengths like red or orange. - Although violet light is scattered slightly more than blue, our eyes are more sensitive to blue, and some violet light is absorbed by the upper atmosphere. As a result, we predominantly perceive the sky as **blue** during the day. At sunrise or sunset, the sky turns red or orange because sunlight passes through more of the atmosphere, scattering away most of the blue and leaving longer wavelengths like red and orange to dominate. 🌅💙 Let me know if you'd like a deeper explanation! 😊 ``` The template is basic and doesn't support tool calling.
GiteaMirror commented 2025-11-12 14:38:26 -06:00
Author
Owner
Copy Link

@logxdx commented on GitHub (Oct 19, 2025):

None of the LFM2 models are good at tool calling.

I created a modelfile for their fine tuned tool use model according to what they say on their HuggingFace page, but THEY FAIL 100% OF THE TIME.

But LFM2 models are very fast and thus good for summarization tasks.

Thanks for the effort of supporting these 🙌

@logxdx commented on GitHub (Oct 19, 2025): **None of the LFM2 models are good at tool calling.** I created a modelfile for their fine tuned tool use model according to what they say on their HuggingFace page, but THEY FAIL 100% OF THE TIME. But LFM2 models are very fast and thus good for summarization tasks. Thanks for the effort of supporting these 🙌
GiteaMirror commented 2025-11-12 14:38:26 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 19, 2025):

Did the modelfile you created enable tool use?

@rick-github commented on GitHub (Oct 19, 2025): Did the modelfile you created enable tool use?
GiteaMirror commented 2025-11-12 14:38:27 -06:00
Author
Owner
Copy Link

@logxdx commented on GitHub (Oct 19, 2025):

Yup, it did show it supports tool calls, but the model NEVER calls the tools.

Check out the official Chat Template

My LFM2 Tool Support Modelfile 
# Modelfile generated by "ollama show"
# To build a new Modelfile based on this, replace FROM with:

FROM LFM2:1.2

TEMPLATE """
<|startoftext|>
{{- if .System -}}
<|im_start|>system
{{- .System -}}
{{-  if .Tools -}}
<|tool_list_start|>
{{ range .Tools -}}
{{- json .Function -}},
{{- end -}}
<|tool_list_end|>
{{ end }}
<|im_end|>
{{- end -}}


{{- range $i, $_ := .Messages }}
{{ $last := eq (len (slice $.Messages $i)) 1 }}

{{- if eq .Role "user" -}}
<|im_start|>user
{{- .Content -}}
<|im_end|>

{{- else if eq .Role "assistant" -}}
<|im_start|>assistant
{{- if .Content -}}
{{- .Content -}}
{{- else if .ToolCalls -}}
<|tool_call_start|>
{{- range .ToolCalls -}}
{- "name": "{{ .Function.Name }}", "arguments": {{ .Function.Arguments }} -}
{{- end -}}
<|tool_call_end|>
{{- end -}}
<|im_end|>

{{- else if eq .Role "tool" -}}
<|im_start|>tool
<|tool_response_start|>
{{- .Content -}}
<|tool_response_end|>
<|im_end|>

{{- end -}}
{{- end -}}
<|im_start|>assistant
"""

PARAMETER stop <|start_of_text|>
PARAMETER stop <|im_start|>
PARAMETER stop <|im_end|>
PARAMETER temperature 0.3
PARAMETER repeat_penalty 1.05
PARAMETER min_p 0.15
PARAMETER num_ctx 32000
@logxdx commented on GitHub (Oct 19, 2025): Yup, it did show it supports tool calls, but the model NEVER calls the tools. Check out the official [Chat Template](https://huggingface.co/LiquidAI/LFM2-1.2B) <details> <summary> My LFM2 Tool Support Modelfile </summary> ```bash # Modelfile generated by "ollama show" # To build a new Modelfile based on this, replace FROM with: FROM LFM2:1.2 TEMPLATE """ <|startoftext|> {{- if .System -}} <|im_start|>system {{- .System -}} {{- if .Tools -}} <|tool_list_start|> {{ range .Tools -}} {{- json .Function -}}, {{- end -}} <|tool_list_end|> {{ end }} <|im_end|> {{- end -}} {{- range $i, $_ := .Messages }} {{ $last := eq (len (slice $.Messages $i)) 1 }} {{- if eq .Role "user" -}} <|im_start|>user {{- .Content -}} <|im_end|> {{- else if eq .Role "assistant" -}} <|im_start|>assistant {{- if .Content -}} {{- .Content -}} {{- else if .ToolCalls -}} <|tool_call_start|> {{- range .ToolCalls -}} {- "name": "{{ .Function.Name }}", "arguments": {{ .Function.Arguments }} -} {{- end -}} <|tool_call_end|> {{- end -}} <|im_end|> {{- else if eq .Role "tool" -}} <|im_start|>tool <|tool_response_start|> {{- .Content -}} <|tool_response_end|> <|im_end|> {{- end -}} {{- end -}} <|im_start|>assistant """ PARAMETER stop <|start_of_text|> PARAMETER stop <|im_start|> PARAMETER stop <|im_end|> PARAMETER temperature 0.3 PARAMETER repeat_penalty 1.05 PARAMETER min_p 0.15 PARAMETER num_ctx 32000 ``` </details>
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
client2
cloud
compatibility
context-length
create
docker
documentation
embeddings
engine
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
install
integration
intel
js
linux
macos
memory
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
windows
wsl
No Label model
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama-ollama#8334
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?