https://www.datasculpting.ai/inference/ Mon, 26 Oct 2026 11:30:30 +0000 MyBB https://www.datasculpting.ai/inference/showthread.php?tid=1415 Sun, 18 Oct 2026 03:38:57 +0000 vectorember]]> https://www.datasculpting.ai/inference/showthread.php?tid=1415
Just sharing a setup that ended up working reasonably well: Mistral, MLX, M3 Max, with a lot of attention on batch size.

Maybe useful as a starting point for newer folks.]]>
Just sharing a setup that ended up working reasonably well: Mistral, MLX, M3 Max, with a lot of attention on batch size.

Maybe useful as a starting point for newer folks.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1404 Sat, 17 Oct 2026 21:52:36 +0000 cacheline]]> https://www.datasculpting.ai/inference/showthread.php?tid=1404
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with INT4.

Curious whether that lines up with what other people are seeing.]]>
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with INT4.

Curious whether that lines up with what other people are seeing.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1398 Sat, 17 Oct 2026 19:37:25 +0000 cacheline]]> https://www.datasculpting.ai/inference/showthread.php?tid=1398
Two paths both seem viable, but they break down differently once quant drift becomes important.

What would you pick if you had to live with it for a few months?

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.]]>
Two paths both seem viable, but they break down differently once quant drift becomes important.

What would you pick if you had to live with it for a few months?

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1402 Sat, 17 Oct 2026 09:04:48 +0000 thetashed24]]> https://www.datasculpting.ai/inference/showthread.php?tid=1402
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with continuous batching.

Maybe obvious to some of you, but it clicked for me this week.]]>
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with continuous batching.

Maybe obvious to some of you, but it clicked for me this week.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1352 Fri, 16 Oct 2026 18:02:57 +0000 phasespark]]> https://www.datasculpting.ai/inference/showthread.php?tid=1352
One thing I keep noticing is that people underestimate long context until they actually try to run this stuff for more than one afternoon.

Curious whether that lines up with what other people are seeing.]]>
One thing I keep noticing is that people underestimate long context until they actually try to run this stuff for more than one afternoon.

Curious whether that lines up with what other people are seeing.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1368 Fri, 16 Oct 2026 07:05:26 +0000 batchsize]]> https://www.datasculpting.ai/inference/showthread.php?tid=1368
I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about MoE serving.

I'm especially interested in mistakes worth avoiding.

This is mostly for local work rather than public serving.]]>
I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about MoE serving.

I'm especially interested in mistakes worth avoiding.

This is mostly for local work rather than public serving.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1256 Fri, 16 Oct 2026 03:04:52 +0000 cacheline]]> https://www.datasculpting.ai/inference/showthread.php?tid=1256
My current setup is LM Studio with B200, and I keep running into MoE serving.

Not looking for the perfect answer, mostly what worked in real use.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.]]>
My current setup is LM Studio with B200, and I keep running into MoE serving.

Not looking for the perfect answer, mostly what worked in real use.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1255 Fri, 16 Oct 2026 02:28:18 +0000 thetaplane]]> https://www.datasculpting.ai/inference/showthread.php?tid=1255
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around MLX and GB200.

Would rather learn from your scars than collect my own all at once.

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.]]>
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around MLX and GB200.

Would rather learn from your scars than collect my own all at once.

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1249 Thu, 15 Oct 2026 04:01:02 +0000 deltapoint17]]> https://www.datasculpting.ai/inference/showthread.php?tid=1249
Just sharing a setup that ended up working reasonably well: FLUX, Ollama, 7900 XTX, with a lot of attention on long context.

Maybe useful as a starting point for newer folks.]]>
Just sharing a setup that ended up working reasonably well: FLUX, Ollama, 7900 XTX, with a lot of attention on long context.

Maybe useful as a starting point for newer folks.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1198 Thu, 15 Oct 2026 00:34:22 +0000 cacheline]]> https://www.datasculpting.ai/inference/showthread.php?tid=1198
This isn't revolutionary, but the combination of PyTorch and M3 Max felt surprisingly usable once I tuned around KV cache pressure.

If someone wants the exact settings I can clean them up and post them.]]>
This isn't revolutionary, but the combination of PyTorch and M3 Max felt surprisingly usable once I tuned around KV cache pressure.

If someone wants the exact settings I can clean them up and post them.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1238 Wed, 14 Oct 2026 21:21:41 +0000 stackpilot]]> https://www.datasculpting.ai/inference/showthread.php?tid=1238
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around MLX and Threadripper box.

I'm especially interested in mistakes worth avoiding.

This is mostly for local work rather than public serving.]]>
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around MLX and Threadripper box.

I'm especially interested in mistakes worth avoiding.

This is mostly for local work rather than public serving.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1226 Wed, 14 Oct 2026 14:40:19 +0000 batchsize]]> https://www.datasculpting.ai/inference/showthread.php?tid=1226
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around vLLM and 4090.

Would rather learn from your scars than collect my own all at once.

I care more about stability than squeezing the last possible tokens/sec out of it.

The part that surprised me was how much the boring details mattered more than the model choice.]]>
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around vLLM and 4090.

Would rather learn from your scars than collect my own all at once.

I care more about stability than squeezing the last possible tokens/sec out of it.

The part that surprised me was how much the boring details mattered more than the model choice.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1188 Wed, 14 Oct 2026 03:56:47 +0000 routearch]]> https://www.datasculpting.ai/inference/showthread.php?tid=1188
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with INT4.

Maybe obvious to some of you, but it clicked for me this week.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.]]>
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with INT4.

Maybe obvious to some of you, but it clicked for me this week.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1111 Mon, 12 Oct 2026 17:42:15 +0000 routearch]]> https://www.datasculpting.ai/inference/showthread.php?tid=1111
I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about prefix caching.

I'm especially interested in mistakes worth avoiding.

This is mostly for local work rather than public serving.]]>
I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about prefix caching.

I'm especially interested in mistakes worth avoiding.

This is mostly for local work rather than public serving.]]> https://www.datasculpting.ai/inference/showthread.php?tid=1119 Mon, 12 Oct 2026 16:35:07 +0000 batchcraft]]> https://www.datasculpting.ai/inference/showthread.php?tid=1119
If you were starting from scratch today for this use case, what would you choose first and what would you ignore?

I'm especially interested in mistakes worth avoiding.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.]]>
If you were starting from scratch today for this use case, what would you choose first and what would you ignore?

I'm especially interested in mistakes worth avoiding.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.]]>