Discussions

• 1 You must be logged in to vote

  💬

  Gemma 4 is now supported only 9 TG tok/s — 1/5 of Qwen 3.5

  gogola2020 started Apr 4, 2026 in General

  

  

  1
• 4 You must be logged in to vote

  💡

  Can we have TurboQuant back?

  kuanjames started Mar 28, 2026 in Ideas

  

  

  

  

  14
• 1 You must be logged in to vote

  💬

  KV Cache Quantization

  ronaldslc started Apr 3, 2026 in General

  

  

  0
• 3 You must be logged in to vote

  🙏

  Gemma4 models not supported

  AndrewY-unique asked Apr 3, 2026 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  Embedding problem in Openclaw

  shiqilin1688 started Apr 2, 2026 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  Are the Intelligence benchmarks overly strict?

  CharlesJS asked Mar 29, 2026 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Duplicate requests from Claude Code

  torarnv asked Mar 29, 2026 in Q&A · Answered

  

  

  2
• 1 You must be logged in to vote

  💬

  Sponsor/support button

  HLGRL started Mar 31, 2026 in General

  

  

  1
• 2 You must be logged in to vote

  💡

  Add more benchmarks

  Revive-Curiosity started Mar 25, 2026 in Ideas

  

  

  

  

  4
• 8 You must be logged in to vote

  💬

  A quick hello and update from the oMLX dev

  jundot started Mar 27, 2026 in General

  

  

  

  

  3
• 1 You must be logged in to vote

  💡

  Feature Request: Consider Integration with flash-moe

  sunjianan9900 started Mar 22, 2026 in Ideas

  

  

  

  

  3
• 1 You must be logged in to vote

  💬

  Is there a system prompt option for basic chat in the omlx chat ui?

  johnlovesgoats started Mar 28, 2026 in General

  

  

  1
• 1 You must be logged in to vote

  💬

  Should I expect caching when dealing with images?

  Muritavo started Mar 26, 2026 in General · Closed

  

  

  1
• 1 You must be logged in to vote

  🙏

  Ternary LM support (BitNet and Falcon)?

  TomLucidor asked Mar 28, 2026 in Q&A · Unanswered

  

  

  0
• 2 You must be logged in to vote

  🙏

  Does oQ allow me to download a bigger model and use oMLX to generate the smaller bit version of it to fit in RAM?

  tty168 asked Mar 25, 2026 in Q&A · Unanswered

  

  

  

  

  5
• 1 You must be logged in to vote

  💬

  Skip V dequant entirely for positions with negligible attention

  mbauer started Mar 27, 2026 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  Spec prefill

  kuanjames asked Mar 27, 2026 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Queen 122B

  lyjmry asked Mar 26, 2026 in Q&A · Unanswered

  

  

  1
• 2 You must be logged in to vote

  💬

  System prompt

  francesshei started Mar 21, 2026 in General

  

  

  1
• 1 You must be logged in to vote

  💬

  Help me understand these inconsistent performance metric with oQ

  eokic started Mar 24, 2026 in General

  

  

  0
• 1 You must be logged in to vote

  💬

  Easy way to have different settings for same model?

  epicycle started Mar 23, 2026 in General

  

  

  0
• 3 You must be logged in to vote

  🙌

  Awesome MLX — curated directory of 80+ MLX ecosystem projects (omlx featured)

  raullenchai started Mar 23, 2026 in Show and tell

  

  

  0
• 2 You must be logged in to vote

  💡

  RDMA / JACCL for distributed inferrence

  blightbow started Mar 21, 2026 in Ideas

  

  

  0
• 1 You must be logged in to vote

  💬

  Vision model issues

  TipKnuckle started Mar 20, 2026 in General

  

  

  0
• 1 You must be logged in to vote

  💬

  Openclaw Model Choices

  zipzagster started Mar 18, 2026 in General

  

  

  2