-
Notifications
You must be signed in to change notification settings - Fork 3
All issues
Issue creation is restricted in this repository
- #24 · rajveer43 opened
on Jul 26, 2026 1
Issues
is:issue state:open
is:issue state:open
Search results
UI: Live KV telemetry & logs — compression, memory, request activity
enhancementNew feature or requestNew feature or requestStatus: Open.#36 In rajveer43/VeloxQuant-MLX;UI: Method & default-profile panel — understand and pick KV compression without Python
documentationImprovements or additions to documentationImprovements or additions to documentationenhancementNew feature or requestNew feature or requestStatus: Open.#35 In rajveer43/VeloxQuant-MLX;UI: Server lifecycle — Start/Stop, network, copyable OpenAI endpoints
enhancementNew feature or requestNew feature or requestStatus: Open.#34 In rajveer43/VeloxQuant-MLX;UX: VeloxQuant control panel (oMLX-like Start/Stop desktop experience)
enhancementNew feature or requestNew feature or requestStatus: Open.#33 In rajveer43/VeloxQuant-MLX;- Status: Open.#31 In rajveer43/VeloxQuant-MLX;
A2ATS-adapted: windowed RoPE rotates far keys (paper leaves them unrotated), b conflated with w, per-step distance gating frozen at write time
bugSomething isn't workingSomething isn't workingdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#29 In rajveer43/VeloxQuant-MLX;RFC: OpenAI-compatible server — mlx_lm.server already serves 35/40 caches, but compression is accounting-only at runtime
enhancementNew feature or requestNew feature or requestStatus: Open.#27 In rajveer43/VeloxQuant-MLX;- Status: Open.#24 In rajveer43/VeloxQuant-MLX;
- Status: Open.#20 In rajveer43/VeloxQuant-MLX;
VeloXQuant Metal Audit Report
enhancementNew feature or requestNew feature or requestStatus: Open.#19 In rajveer43/VeloxQuant-MLX;Integration: bridge oMLX's KV-cache quantizer flag into VeloxQuant-MLX (35 methods)
enhancementNew feature or requestNew feature or requestStatus: Open.#5 In rajveer43/VeloxQuant-MLX;LM Studio integration: run VeloxQuant-MLX KV-cache compression inside LM Studio's MLX engine
enhancementNew feature or requestNew feature or requestStatus: Open.#4 In rajveer43/VeloxQuant-MLX;