4 ms·

what models are you using on that? My experiences with apple hardware have convinced me that it is not really good enough for coding locally.

irishcoffee 5mo ago
It isn’t the models, it’s the closed api and the tooling associated with it. It’s driving me crazy how not-talked-about this is.
- datadrivenangel 5mo ago
  As in the coding harnesses?
  - irishcoffee 5mo ago
    If I could leverage the same closed api VSCode uses, the entire moat is drained. What you call harnesses I call… bullshit?
- klausa 5mo ago
  You can point both Codex and Claude Code at a local model and they'll work just fine. Codex even explicitly supports that as a feature! [1] With a nice UI on top, for the desktop app too: [2] [1]: https://developers.openai.com/codex/config-advanced#custom-model-providers https://developers.openai.com/codex/config-advanced#custom-m... [2]: https://docs.ollama.com/integrations/codex-app https://docs.ollama.com/integrations/codex-app
girvo 5mo ago
DeepSeek v4 Flash, various quantised versions of Kimi K2.6, MiniMax 2.7, Qwen 3.5 “full sized, with a dual spark setup you can fit some decent setups on here My single spark has me running Qwen 3.6 27B and antirez’s specially quantised DeepSeek v4 Flash (which is shockingly impressive)
- Tepix 5mo ago
  Kimi K2.6 does not run well on 256GB.
  - girvo 5mo ago
    True, I might be thinking of some of the communities four-Spark clusters for it; it’s already int4 right?
    - Tepix 5mo ago
      Yeah, the default quants are 595GB. Even four Sparks would require a quant lower than 4bit
  - zozbot234 5mo ago
    Have you tried it? It would be slow for sure, but the main limitation AIUI would actually be storing the context in RAM - models like Kimi and GLM have high demands there which limit your ability to get meaningful aggregate throughput via large batches.
    - Tepix 5mo ago
      No need to try really. 1100b weights with 256GB RAM that‘s less than 1.8 bits per weight if you want a little bit of context. How is that supposed to give good results?
mv4 5mo ago
Minimax M2.7.

irishcoffee 5mo ago

It isn’t the models, it’s the closed api and the tooling associated with it. It’s driving me crazy how not-talked-about this is.

datadrivenangel 5mo ago
As in the coding harnesses?
- irishcoffee 5mo ago
  If I could leverage the same closed api VSCode uses, the entire moat is drained. What you call harnesses I call… bullshit?
klausa 5mo ago
You can point both Codex and Claude Code at a local model and they'll work just fine. Codex even explicitly supports that as a feature! [1] With a nice UI on top, for the desktop app too: [2] [1]: https://developers.openai.com/codex/config-advanced#custom-model-providers https://developers.openai.com/codex/config-advanced#custom-m... [2]: https://docs.ollama.com/integrations/codex-app https://docs.ollama.com/integrations/codex-app

girvo 5mo ago

DeepSeek v4 Flash, various quantised versions of Kimi K2.6, MiniMax 2.7, Qwen 3.5 “full sized, with a dual spark setup you can fit some decent setups on here My single spark has me running Qwen 3.6 27B and antirez’s specially quantised DeepSeek v4 Flash (which is shockingly impressive)

Tepix 5mo ago
Kimi K2.6 does not run well on 256GB.
- girvo 5mo ago
  True, I might be thinking of some of the communities four-Spark clusters for it; it’s already int4 right?
  - Tepix 5mo ago
    Yeah, the default quants are 595GB. Even four Sparks would require a quant lower than 4bit
- zozbot234 5mo ago
  Have you tried it? It would be slow for sure, but the main limitation AIUI would actually be storing the context in RAM - models like Kimi and GLM have high demands there which limit your ability to get meaningful aggregate throughput via large batches.
  - Tepix 5mo ago
    No need to try really. 1100b weights with 256GB RAM that‘s less than 1.8 bits per weight if you want a little bit of context. How is that supposed to give good results?

mv4 5mo ago

Minimax M2.7.