Search results

check_batch_size_fields rejects configs that rely on the documented micro_batch_size/gradient_accumulation_steps defaults

Status: Open.

#3836 In axolotl-ai-cloud/axolotl;

· ErenAta16 opened on Jul 12, 2026

DPO chat_template Strategy Drops tool_calls and Never Passes tools to the Chat Template

Status: Open.

#3824 In axolotl-ai-cloud/axolotl;

· inoutro opened on Jul 10, 2026

[ Maintainability improvements ] Quack 5 -> 6

enhancementNew feature or requestNew feature or request

Status: Open.

#3823 In axolotl-ai-cloud/axolotl;

· NanoCode012 opened on Jul 10, 2026

[ Discussion ] NVFP4 LoRA Merge

enhancementNew feature or requestNew feature or request

help wantedExtra attention is neededExtra attention is needed

Status: Open.

#3803 In axolotl-ai-cloud/axolotl;

· NanoCode012 opened on Jul 8, 2026

PEFT bump pending

enhancementNew feature or requestNew feature or request

Status: Open.

#3801 In axolotl-ai-cloud/axolotl;

· thad0ctor opened on Jul 5, 2026

Expert-granularity CPU offload for quantized MoE: 30B-A3B QLoRA in ~7 GB (single resident expert layer, extends the layer_offloading idea)

waiting for reporter

Status: Open.

#3787 In axolotl-ai-cloud/axolotl;

· pjordanandrsn opened on Jul 2, 2026

merge-lora fails on NVFP4 scattermoe MoE (glm_moe_dsa): memory_efficient -> tensor shape mismatch; legacy -> save_pretrained 'invalid python storage'

waiting for reporter

Status: Open.

#3773 In axolotl-ai-cloud/axolotl;

· trevorgordon981 opened on Jul 1, 2026

chat_template Strategy Never Trains the Gemma Turn Terminator

bugSomething isn't workingSomething isn't working

Status: Open.

#3754 In axolotl-ai-cloud/axolotl;

· thad0ctor opened on Jun 25, 2026

Claude /skills exploration

enhancementNew feature or requestNew feature or request

Status: Open.

#3749 In axolotl-ai-cloud/axolotl;

· thad0ctor opened on Jun 23, 2026

Add RLOO and REINFORCE++ Advantage Estimators as GRPO Extensions

enhancementNew feature or requestNew feature or request

Status: Open.

#3676 In axolotl-ai-cloud/axolotl;

· ved1beta opened on May 22, 2026

Add NSA/FSA Native Sparse Attention Support for DeepSeek / Kimi Linear Models

enhancementNew feature or requestNew feature or request

Status: Open.

#3675 In axolotl-ai-cloud/axolotl;

· ved1beta opened on May 22, 2026

Security Advisory Advisory

bugSomething isn't workingSomething isn't working

Status: Open.

#3642 In axolotl-ai-cloud/axolotl;

· beanduan22 opened on May 4, 2026