…fallbacks

1. layers.py: fix RowParallelLinearWithLoRA bias handling to pass bias
   into quant_method.apply(), matching base RowParallelLinear behavior;
   add interleaved gate/up layout support in FusedMoEWithLoRA for models
   using gemm1_alpha (e.g. gpt-oss-20b)

2. mem_pool.py: zero-initialize all LoRA buffers (torch.empty ->
   torch.zeros) to prevent garbage values in unused slots

3. utils.py: fall back to config.intermediate_size when
   moe_intermediate_size is not available in get_hidden_dim (supports
   GptOss, Mixtral, OLMoE, PhiMoE, GraniteMoE, Grok, etc.); accept
   PEFT shorthand "all-linear" in get_normalized_target_modules; fix
   isinstance order in auto_detect_lora_target_modules so ParallelLMHead
   is checked before VocabParallelEmbedding

4. gpt_oss.py: add should_apply_lora() to GptOssForCausalLM for
   explicit LoRA module filtering, consistent with Qwen3VLMoe

Made-with: Cursor

Regression test comparing SGLang LoRA logprobs against reference
training logprobs (KL threshold 1e-2). Uses 8-GPU H200 suite with
triton MoE runner and shared outer LoRA mode.

Adapter checkpoint: yushengsu/lora-diff-gpt-oss-20b

Made-with: Cursor

Pre-allocate MoE intermediate buffers before memory profiling so
KV cache sizing accounts for them. Reuse fixed buffers during
capture/replay instead of dynamic torch.empty() allocations.

…raph

Extract get_triton_quant_info() into FusedMoEMethodBase and each quant
method (Fp8, W8A8Fp8, W8A8Int8, BlockInt8, MoeWNA16, Unquantized) so
FusedMoEWithLoRA uses the polymorphic method instead of hardcoding
TritonMoeQuantInfo. Enables LoRA on quantized MoE models.

Made-with: Cursor

- Add ReplicatedLinearWithLoRA for fused_qkv_a_proj_with_mqa, applying
  LoRA B via two separate sgemm calls for unequal output partitions
  (q_a_proj=1536 vs kv_a_proj_with_mqa=576). B slices are precomputed
  in set_lora_info to avoid per-forward allocation.
- Add normalize_fused_qkv_a_proj to fuse q_a_proj + kv_a_proj_with_mqa
  adapter weights into a single stacked entry.
- Add stack_num parameter to run_lora_a_sgemm across all 3 backends.
- Fix o_proj hidden dim to use v_head_dim for MLA models.
- Fix gate_up_proj/down_proj hidden dim to use per-layer shared expert
  intermediate size on MoE layers.
- Exclude ReplicatedLinear from TP sharding in memory pool allocation.

Made-with: Cursor

Made-with: Cursor

- Force triton-compatible MoE weights when LoRA is enabled for compressed
  tensors quantized models (avoid Marlin path which is incompatible)
- Refactor get_triton_quant_info into a reusable method for LoRA MoE runner
- Make MoE LoRA runner backend detection robust with hasattr fallback
- Handle multi-modal model configs via get_text_config() in LoRAManager
- Add CI test for Kimi-K2.5 LoRA logprob accuracy

Made-with: Cursor