(ollama.com)

Show context

buyucu[dead post] ◴[16 May 25 07:19 UTC] No.44002611[source]▶

[flagged]

1. hexmiles ◴[16 May 25 10:26 UTC] No.44003680[source]▶

If I understood it correctly: this time no, it is actually new engine builded by the ollama team indipendent from llama.cpp

2. Havoc ◴[16 May 25 10:37 UTC] No.44003746[source]▶

llama.cpp added support for vision 6 days ago.

See SimonW post here:

>If I understood it correctly

You understood it exactly like they wanted you to...

3. buyucu ◴[16 May 25 10:37 UTC] No.44003749[source]▶

I doubt it. Llama.cpp just added support for the same models a few weeks ago. Folks at ollama just did a git pull.

4. magicalhippo ◴[16 May 25 11:28 UTC] No.44004061[source]▶

It's open source, you could have checked. Seems indeed like the new engine cuts out llama.cpp, using GGML libary directly.

replies(1): >>44005517 #

5. ◴[16 May 25 11:35 UTC] No.44004120[source]▶

6. buyucu ◴[16 May 25 13:49 UTC] No.44005517{3}[source]▶

seriously? who do you think develops ggml?

hint: it's llama.cpp

Ollama's new engine for multimodal models