Docker version of llama.cpp is not detected

[fessmm]

Hi, i searched for it but could not find any issue describing exactly my problem.

When running ollama with docker compose -> llmfit detects ollama
When running llama.cpp with docker compose -> llmfit cannot detect llama.cpp

Ports are default:

• Ollama -> 11434
• llama.cpp -> 8080

If you need anything else, just let me know.
Thanks!

[openbrian]

I tested this using an alternate port. It worked.

LLAMA_CPP_HOST=localhost:8090 ./target/release/llmfit

[stepbystepcode]

I tested this using an alternate port. It worked.

LLAMA_CPP_HOST=localhost:8090 ./target/release/llmfit

not work for me.

LLAMA_CPP_HOST=localhost:8001 llmfit still cannot find llama.cpp backend.
podman run --rm -it
--security-opt=label=disable
--device nvidia.com/gpu=all
-v /home/stepbystep/Documents/models:/models:z
-p 8001:8001
ghcr.io/ggml-org/llama.cpp:server-cuda
-m /models/Qwen3.5-27B-GGUF/Qwen3.5-27B-UD-Q4_K_XL.gguf
--alias "unsloth/Qwen3.5-27B"
--port 8001
--host 0.0.0.0
--temp 0.6
--top-p 0.95
--top-k 20
--min-p 0.00
--cache-type-k q8_0 --cache-type-v q8_0
--flash-attn on --fit on
--ctx-size 81920

environment:
OS: Fedora Linux 43 (Workstation Edition) x86_64
Host: MS-7D76 (2.0)
Kernel: Linux 6.18.12-200.fc43.x86_64
CPU: AMD Ryzen 9 7900X (24) @ 5.74 GHz
GPU 1: NVIDIA GeForce RTX 3090 Ti [Discrete]
GPU 2: AMD Raphael [Integrated]

[openbrian]

Sorry, I meant I tested a patch for this on a fork.