Model configuration does not work (Error: Model llama2 not found)

[Sam]

When configuring the Ollama model to use loz does not seem to actually save the change and crashes out:

samm-mbp ~ $ ollama pull tinyllama:1.1b-chat-v1-fp16
pulling manifest
pulling 659305c3a0fd... 100% ...
▕███████████████████████████████████████████████████████████████████████████████████████▏  2.2 GB
verifying sha256 digest
writing manifest
removing any unused layers
success

samm-mbp ~ $ loz config model tinyllama:1.1b-chat-v1-fp16
llama-node/lib/ollama.js:80
            throw new Error(`Model ${model} not found.`);
                  ^

Error: Model llama2 not found.
    at Ollama.setModel (loz/node_modules/ollama-node/lib/ollama.js:80:19)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)

Node.js v20.9.0

samm-mbp ~ [1] $ loz
Loz: a simple CLI for LLM
  api: ollama
  model: llama2

> config model tinyllama:1.1b-chat-v1-fp16
The model has been updated to 'tinyllama:1.1b-chat-v1-fp16'

> write a fake commit message about cats
loz/node_modules/ollama-node/lib/ollama.js:80
            throw new Error(`Model ${model} not found.`);
                  ^

Error: Model llama2 not found.
    at Ollama.setModel (loz/node_modules/ollama-node/lib/ollama.js:80:19)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)

Node.js v20.9.0

On the Ollama server side I see Loz hitting two API endpoints:

ollama serve
time=2024-02-15T07:40:31.676+11:00 level=INFO source=images.go:863 msg="total blobs: 60"
time=2024-02-15T07:40:31.679+11:00 level=INFO source=images.go:870 msg="total unused blobs removed: 0"
time=2024-02-15T07:40:31.681+11:00 level=INFO source=routes.go:999 msg="Listening on 127.0.0.1:11434 (version 0.1.24)"
time=2024-02-15T07:40:31.681+11:00 level=INFO source=payload_common.go:106 msg="Extracting dynamic libraries..."
time=2024-02-15T07:40:31.700+11:00 level=INFO source=payload_common.go:145 msg="Dynamic LLM libraries [metal]"
[GIN] 2024/02/15 - 07:40:39 | 200 |      29.458µs |       127.0.0.1 | GET      "/api/version"
[GIN] 2024/02/15 - 07:40:47 | 200 |   13.131041ms |       127.0.0.1 | GET      "/api/tags"

[Joone Hur]

Loz currently supports only Llama2 and CodeLlama. I will consider supporting any LLM models that Ollama supports.

[Sam]

Oh really, llama2 is getting pretty far behind other models these days, perhaps it could just support loading from the Ollama model list?

[Joone Hur]

It is hard-coded. Okay, I will fix it.

[Joone Hur]

This bug has been resolved in the latest release. Please let me know if the issue reoccurs.