Search results

error occurred when deploying the transformed model using the vllm server

Status: Open.

#40 In MuLabPKU/TransMLA;

· Wfd567 opened on Oct 27, 2025

Errors when running target model

Status: Open.

#39 In MuLabPKU/TransMLA;

· Laulava-Velho opened on Oct 21, 2025

Support Qwen3, e.g Qwen3-32B(GQA) can be tranasformed to MLA

Status: Open.

#38 In MuLabPKU/TransMLA;

· xueliangyang-oeuler opened on Oct 13, 2025

Question: Applicability of TransMLA to DLM

Status: Open.

#37 In MuLabPKU/TransMLA;

· VIROBO-15 opened on Oct 8, 2025

Question: Details of KV cache compression ratios and settings for all models

Status: Open.

#36 In MuLabPKU/TransMLA;

· bg51717 opened on Sep 21, 2025

Error with running qwen-3b-instruct converted

Status: Open.

#35 In MuLabPKU/TransMLA;

· nikita-yatchenko opened on Sep 7, 2025

No performance difference with vLLM: qwen2.5-7b vs qwen2.5-7b-MLA

Status: Open.

#34 In MuLabPKU/TransMLA;

· nikita-yatchenko opened on Sep 4, 2025

Can't run the converted model with vllm.LLM

Status: Open.

#32 In MuLabPKU/TransMLA;

· ItGirls opened on Aug 28, 2025

Release TransMLA converted checkpoints on Hugging Face

Status: Open.

#31 In MuLabPKU/TransMLA;

· NielsRogge opened on Aug 26, 2025

is support gpt-oss ?

Status: Open.

#30 In MuLabPKU/TransMLA;

· libai-lab opened on Aug 23, 2025

add benchmark

Status: Open.

#25 In MuLabPKU/TransMLA;

· huuyu opened on Jun 30, 2025

qwen2-7b and llama2-7b unable to deploy

Status: Open.

#24 In MuLabPKU/TransMLA;

· mingyangHao opened on Jun 30, 2025