-
Notifications
You must be signed in to change notification settings - Fork 174
All issues
Issue creation is restricted in this repository
- #2283 · XuehaoSun opened
on Sep 3, 2026 3
Issues
is:issue state:open
is:issue state:open
Search results
[Bug]: GLM5.3 some weights should kept as FP32
bugSomething isn't workingSomething isn't workingStatus: Open.[Bug]: MTP logic is unaware of Transformers' layer renaming and quantizes the entire model again
bugSomething isn't workingSomething isn't workingStatus: Open.[Feature]: very slow to quatnize qwen3.8-flash-next for full attention layers
enhancementNew feature or requestNew feature or requestStatus: Open.#2323 In intel/auto-round;[Bug]: Check whether the warning for Gemma is expected
bugSomething isn't workingSomething isn't workingStatus: Open.[Regression]:
auto-round xxx --format auto_round,fakefake mode'sl config.json is a quantized versionbugSomething isn't workingSomething isn't workingStatus: Open.#2310 In intel/auto-round;[Feature]: support data parallel
enhancementNew feature or requestNew feature or requestStatus: Open.RFC: Recurrent Residual Quantization (RRQ) INT2+2+2+2
enhancementNew feature or requestNew feature or requestStatus: Open.#2300 In intel/auto-round;[Feature]: expand AWQ experiments to low-bit INT and Hadamard combinations
enhancementNew feature or requestNew feature or requestStatus: Open.#2298 In intel/auto-round;[Bug]: CPU quantization can't get proper model size
bugSomething isn't workingSomething isn't workingStatus: Open.#2293 In intel/auto-round;[Feature]: optimize moe tuning cost
enhancementNew feature or requestNew feature or requestStatus: Open.#2290 In intel/auto-round;- Status: Open.
[Bug]: Fix AutoRound gradients for Kimi-VL routed experts- #2097
bugSomething isn't workingSomething isn't workingStatus: Open.