09-14-2026, 05:14 AM
I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at MoE serving. Not glamorous advice, but repeatable baselines save a lot of time.
|
Help with Embeddings: early impressions
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
Help with Embeddings: early impressions - by vectorshed43 - 09-12-2026, 03:53 AM
Re: Help with Embeddings: early impressions - by vectorshed43 - 09-13-2026, 04:12 AM
Re: Help with Embeddings: early impressions - by edgecoast - 09-13-2026, 04:58 PM
Re: Help with Embeddings: early impressions - by vectorshed43 - 09-14-2026, 02:18 PM
Re: Help with Embeddings: early impressions - by quantnest - 09-14-2026, 05:14 AM
Re: Help with Embeddings: early impressions - by paperport64 - 09-14-2026, 06:06 AM
Re: Help with Embeddings: early impressions - by vramscope - 09-14-2026, 02:52 PM
Re: Help with Embeddings: early impressions - by quantnest - 09-14-2026, 03:30 PM
Re: Help with Embeddings: early impressions - by ggufwire22 - 09-14-2026, 03:36 PM
Re: Help with Embeddings: early impressions - by vramscope - 09-14-2026, 04:10 PM
|