-
Notifications
You must be signed in to change notification settings - Fork 6
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
docs: add decision guide for upstream native vLLM vs turboquant-vllm
documentationImprovements or additions to documentationImprovements or additions to documentationpriority: P2Medium — quality/researchMedium — quality/researchStatus: Open.#90 In Alberto-Codes/turboquant-vllm;research(rotation): prototype WHT/Hadamard rotation path
enhancementNew feature or requestNew feature or requestpriority: P2Medium — quality/researchMedium — quality/researchStatus: Open.#89 In Alberto-Codes/turboquant-vllm;feat(verify): architecture-aware compatibility and policy recommendations
documentationImprovements or additions to documentationImprovements or additions to documentationenhancementNew feature or requestNew feature or requestpriority: P1High — needed for model support parityHigh — needed for model support parityStatus: Open.#88 In Alberto-Codes/turboquant-vllm;docs: reposition project around HF reference + validation workflow
documentationImprovements or additions to documentationImprovements or additions to documentationpriority: P1High — needed for model support parityHigh — needed for model support parityStatus: Open.#87 In Alberto-Codes/turboquant-vllm;meta: reposition turboquant-vllm around HF reference, verification, and research
documentationImprovements or additions to documentationImprovements or additions to documentationenhancementNew feature or requestNew feature or requestpriority: P1High — needed for model support parityHigh — needed for model support parityStatus: Open.#86 In Alberto-Codes/turboquant-vllm;test(api): add public API surface and HF-without-vLLM tests
enhancementNew feature or requestNew feature or requestpriority: P3Low — housekeepingLow — housekeepingtestTest improvements and additionsTest improvements and additionsStatus: Open.#23 In Alberto-Codes/turboquant-vllm;test(vllm): add GPU integration test with real vLLM engine
enhancementNew feature or requestNew feature or requestpriority: P3Low — housekeepingLow — housekeepingtestTest improvements and additionsTest improvements and additionsStatus: Open.#22 In Alberto-Codes/turboquant-vllm;chore(tests): add defensive .clone() for shared tensor in _make_impl()
choreMaintenance and cleanupMaintenance and cleanuppriority: P3Low — housekeepingLow — housekeepingtestTest improvements and additionsTest improvements and additionsStatus: Open.#18 In Alberto-Codes/turboquant-vllm;chore(tests): add
from __future__ import annotationsto 3 test fileschoreMaintenance and cleanupMaintenance and cleanuppriority: P3Low — housekeepingLow — housekeepingtestTest improvements and additionsTest improvements and additionsStatus: Open.#17 In Alberto-Codes/turboquant-vllm;