project:

project: type: website pre-render:

• docs/scripts/generate_config_docs.py
• docs/scripts/generate_examples_docs.py

quartodoc: dir: docs/api package: axolotl title: API Reference parser: google

sections: - title: Core desc: Core functionality for training contents: - train - evaluate - datasets - convert - prompt_tokenizers - prompters - processing_strategies - logging_config - core.builders.base - core.builders.causal - core.builders.rl - core.training_args - core.training_args_base - core.chat.messages - core.chat.format.chatml - core.chat.format.llama3x - core.chat.format.shared - core.datasets.chat - core.datasets.transforms.chat_builder - title: CLI desc: Command-line interface contents: - cli.main - cli.train - cli.evaluate - cli.args - cli.art - cli.checks - cli.config - cli.delinearize_llama4 - cli.inference - cli.merge_lora - cli.merge_sharded_fsdp_weights - cli.preprocess - cli.quantize - cli.vllm_serve - cli.agent_docs - cli.cloud - cli.cloud.base - cli.cloud.baseten - cli.cloud.modal_ - cli.utils - cli.utils.args - cli.utils.diffusion - cli.utils.fetch - cli.utils.load - cli.utils.lora_merge - cli.utils.sweeps - cli.utils.train - title: Trainers desc: Training implementations contents: - core.trainers.base - core.trainers.constants - core.trainers.trl - core.trainers.mamba - core.trainers.dpo.args - core.trainers.dpo.trainer - core.trainers.ebft - core.trainers.ebft.args - core.trainers.ebft.kernels - core.trainers.ebft.rewards - core.trainers.ebft.strided - core.trainers.ebft.trainer - core.trainers.grpo - core.trainers.grpo.args - core.trainers.grpo.trainer - core.trainers.grpo.async_trainer - core.trainers.grpo.fast_async_trainer - core.trainers.grpo.replay_buffer - core.trainers.grpo.sampler - core.trainers.utils - title: Model Loading desc: Functionality for loading and patching models, tokenizers, etc. contents: - loaders.model - loaders.tokenizer - loaders.processor - loaders.adapter - loaders.patch_manager - loaders.constants - loaders.utils - title: Mixins desc: Mixin classes for augmenting trainers contents: - core.trainers.mixins.activation_checkpointing - core.trainers.mixins.checkpoints - core.trainers.mixins.distributed_parallel - core.trainers.mixins.layer_offloading - core.trainers.mixins.optimizer - core.trainers.mixins.packing - core.trainers.mixins.rng_state_loader - core.trainers.mixins.scheduler - title: Context Managers desc: Context managers for altering trainer behaviors contents: - utils.ctx_managers.sequence_parallel - title: Prompt Strategies desc: Prompt formatting strategies contents: - prompt_strategies.base - prompt_strategies.chat_template - prompt_strategies.alpaca_chat - prompt_strategies.alpaca_instruct - prompt_strategies.alpaca_w_system - prompt_strategies.user_defined - prompt_strategies.llama2_chat - prompt_strategies.completion - prompt_strategies.context_qa - prompt_strategies.creative_acr - prompt_strategies.input_output - prompt_strategies.pretrain - prompt_strategies.stepwise_supervised - prompt_strategies.metharme - prompt_strategies.orcamini - prompt_strategies.pygmalion - prompt_strategies.messages.chat - prompt_strategies.ebft.ebft_chat_multiturn - prompt_strategies.ebft.ebft_opencode - prompt_strategies.ebft.ebft_reasoning - prompt_strategies.ebft.ebft_strided_chat - prompt_strategies.ebft.ebft_strided_structured - prompt_strategies.dpo.chat_template - prompt_strategies.dpo.llama3 - prompt_strategies.dpo.chatml - prompt_strategies.dpo.zephyr - prompt_strategies.dpo.user_defined - prompt_strategies.dpo.passthrough - prompt_strategies.kto.llama3 - prompt_strategies.kto.chatml - prompt_strategies.kto.user_defined - prompt_strategies.orpo.chat_template - prompt_strategies.bradley_terry.chat_template - prompt_strategies.bradley_terry.llama3 - title: Kernels desc: Low-level performance optimizations contents: - kernels.lora - kernels.dora - kernels.geglu - kernels.swiglu - kernels.quantize - kernels.autotune_telemetry - kernels.gemma4_fused_rope - kernels.rms_norm_gated - kernels.utils - title: Monkey Patches desc: Runtime patches for model optimizations contents: - monkeypatch.llama_attn_hijack_flash - monkeypatch.llama_attn_hijack_xformers - monkeypatch.mistral_attn_hijack_flash - monkeypatch.multipack - monkeypatch.relora - monkeypatch.lora_kernels - monkeypatch.utils - monkeypatch.btlm_attn_hijack_flash - monkeypatch.stablelm_attn_hijack_flash - monkeypatch.transformers_fa_utils - monkeypatch.data.batch_dataset_fetcher - monkeypatch.mixtral - monkeypatch.gradient_checkpointing.offload_cpu - monkeypatch.gradient_checkpointing.offload_disk - monkeypatch.deepspeed_utils - monkeypatch.fsdp2_qlora - monkeypatch.gemma4_hybrid_mask - monkeypatch.gemma4_loss_kwargs - monkeypatch.kernelize_fixes - monkeypatch.moe_quant - monkeypatch.scaled_softmax_attn - monkeypatch.torchao_optim - monkeypatch.trainer_accelerator_args - monkeypatch.accelerate.fsdp2 - monkeypatch.accelerate.parallelism_config - monkeypatch.attention.flash_attn_4 - monkeypatch.attention.flex_attn - monkeypatch.attention.fp8_attn - monkeypatch.attention.sage_attn - monkeypatch.attention.xformers - monkeypatch.loss.chunked - monkeypatch.loss.eaft - monkeypatch.models.apertus.activation - monkeypatch.models.falcon_h1.modeling - monkeypatch.models.gemma4_unified.fused_attn - monkeypatch.models.granitemoehybrid.modeling - monkeypatch.models.kimi_linear.patch_kimi_linear - monkeypatch.models.llama4.modeling - monkeypatch.models.mamba_utils - monkeypatch.models.mistral3.mistral_common_tokenizer - monkeypatch.models.nemotron_h.modeling - monkeypatch.models.pixtral.modeling_flash_attention_utils - monkeypatch.models.qwen3.fused_attn - monkeypatch.models.qwen3_5.fused_attn - monkeypatch.models.qwen3_5.modeling - monkeypatch.models.qwen3_5_moe.fused_attn - monkeypatch.models.qwen3_moe.fused_attn - monkeypatch.models.qwen3_next.modeling - monkeypatch.models.qwen3_vl.fused_attn - monkeypatch.models.voxtral.modeling - monkeypatch.peft.utils - monkeypatch.ring_attn.adapters.batch - monkeypatch.ring_attn.patch - monkeypatch.tiled_mlp.base - monkeypatch.tiled_mlp.patch - monkeypatch.trainer.lr - monkeypatch.trainer.trl - monkeypatch.trainer.trl_vllm - monkeypatch.trainer.utils - monkeypatch.transformers.trainer_loss_calc - monkeypatch.xformers_ - title: Utils desc: Utility functions contents: - utils.tokenization - utils.chat_templates - utils.chat_templates.base - utils.lora - utils.model_shard_quant - utils.bench - utils.comet_ - utils.config - utils.cuda13 - utils.datasets - utils.environment - utils.fp32_norms - utils.freeze - utils.import_helper - utils.logging - utils.mlflow_ - utils.tee - utils.trackio_ - utils.train - utils.trainer - utils.wandb_ - utils.weight_serde - utils.schedulers - utils.distributed - utils.dict - utils.generation.sft - utils.mistral.mistral3_processor - utils.mistral.mistral_tokenizer - utils.optimizers.adopt - utils.optimizers.qgalore - utils.data.streaming - utils.data.sft - utils.data.rl - utils.data.lock - utils.data.utils - utils.data.wrappers - utils.quantization - title: Schemas desc: Pydantic data models for Axolotl config contents: - utils.schemas.config - utils.schemas.model - utils.schemas.training - utils.schemas.datasets - utils.schemas.peft - utils.schemas.trl - utils.schemas.multimodal - utils.schemas.integrations - utils.schemas.deprecated - utils.schemas.dynamic_checkpoint - utils.schemas.fsdp - utils.schemas.quantization - utils.schemas.validation - utils.schemas.vllm - utils.schemas.enums - utils.schemas.utils - title: Integrations desc: Third-party integrations and extensions contents: - integrations.base - integrations.config - integrations.cut_cross_entropy - integrations.cut_cross_entropy.args - integrations.densemixer.args - integrations.densemixer.plugin - integrations.diffusion.args - integrations.diffusion.callbacks - integrations.diffusion.generation - integrations.diffusion.plugin - integrations.diffusion.trainer - integrations.diffusion.utils - integrations.expert_parallel.args - integrations.expert_parallel.buffer - integrations.expert_parallel.experts_fn - integrations.expert_parallel.plugin - integrations.expert_parallel.shard - integrations.grokfast.args - integrations.grokfast.optimizer - integrations.hatchery.args - integrations.hatchery.data - integrations.hatchery.plugin - integrations.hatchery.rewards.math_reward - integrations.hatchery.rl_trainer - integrations.hatchery.trainer - integrations.kd - integrations.kd.args - integrations.kd.callbacks - integrations.kd.chat_template - integrations.kd.collator - integrations.kd.collator_online_teacher - integrations.kd.kernels.liger - integrations.kd.topk_logprob.forward_kl - integrations.kd.trainer - integrations.kd.utils - integrations.kernels.args - integrations.kernels.autotune_callback - integrations.kernels.autotune_collector - integrations.kernels.constants - integrations.kernels.plugin - integrations.liger.args - integrations.liger.plugin - integrations.liger.utils - integrations.liger.models.base - integrations.liger.models.deepseekv2 - integrations.liger.models.jamba - integrations.liger.models.llama4 - integrations.liger.models.qwen3 - integrations.liger.models.qwen3_5 - integrations.liger.models.qwen3_5_moe - integrations.liger.models.qwen3_moe - integrations.llm_compressor.args - integrations.llm_compressor.plugin - integrations.llm_compressor.utils - integrations.lm_eval.args - integrations.lm_eval.cli - integrations.mora.args - integrations.mora.plugin - integrations.nemo_gym.args - integrations.nemo_gym.data_producer - integrations.nemo_gym.dataset - integrations.nemo_gym.multi_turn - integrations.nemo_gym.plugin - integrations.nemo_gym.rewards - integrations.nemo_gym.server - integrations.spectrum - integrations.spectrum.args - integrations.swanlab.args - integrations.swanlab.callbacks - integrations.swanlab.completion_logger - integrations.swanlab.plugins - title: Common desc: Common utilities and shared functionality contents: - common.architectures - common.const - common.datasets - title: Models desc: Custom model implementations contents: - models.mamba.configuration_mamba - models.mamba.modeling_mamba - title: Data Processing desc: Data processing utilities contents: - utils.collators.core - utils.collators.batching - utils.collators.dpo - utils.collators.mamba - utils.collators.mm_chat - utils.samplers.multipack - utils.samplers.utils - title: Callbacks desc: Training callbacks contents: - utils.callbacks - utils.callbacks.perplexity - utils.callbacks.profiler - utils.callbacks.lisa - utils.callbacks.mlflow_ - utils.callbacks.comet_ - utils.callbacks.qat - utils.callbacks.dynamic_checkpoint - utils.callbacks.generation - utils.callbacks.models - utils.callbacks.opentelemetry - utils.callbacks.swanlab - utils.callbacks.tokens_per_second - utils.callbacks.trackio_ - title: Scripts desc: Standalone helper scripts contents: - scripts.process_cleanup - scripts.vllm_serve_lora - scripts.vllm_worker_ext - title: Telemetry desc: Usage telemetry contents: - telemetry.callbacks - telemetry.errors - telemetry.manager - telemetry.runtime_metrics website: title: "Axolotl" description: "We make fine-tuning accessible, scalable, and fun" favicon: favicon.jpg

google-analytics: "G-9KYCVJBNMQ"

navbar: logo: image/axolotl_logo_digital_white.svg title: false background: dark pinned: false collapse: false tools: - icon: twitter href: https://twitter.com/axolotl_ai - icon: github href: https://github.com/axolotl-ai-cloud/axolotl/ - icon: discord href: https://discord.gg/7m9sfhzaf3

sidebar: pinned: true collapse-level: 2 style: docked contents: - text: Home href: index.qmd

    - section: "Getting Started"
      contents:
        - docs/getting-started.qmd
        - docs/choosing_method.qmd
        - docs/installation.qmd
        - docs/inference.qmd
        - section: "Model Guides"
          contents:
            - docs/models/kimi-linear.qmd
            - docs/models/plano.qmd
            - docs/models/mimo.qmd
            - docs/models/internvl3_5.qmd
            - docs/models/olmo3.qmd
            - docs/models/trinity.qmd
            - docs/models/arcee.qmd
            - section: "Ministral3"
              contents:
                - docs/models/ministral3.qmd
                - docs/models/ministral3/think.qmd
                - docs/models/ministral3/vision.qmd
            - section: "Magistral"
              contents:
                - docs/models/magistral.qmd
                - docs/models/magistral/think.qmd
                - docs/models/magistral/vision.qmd
            - docs/models/ministral.qmd
            - docs/models/mistral-small.qmd
            - docs/models/voxtral.qmd
            - docs/models/devstral.qmd
            - docs/models/mistral.qmd
            - docs/models/llama-4.qmd
            - docs/models/llama-2.qmd
            - docs/models/qwen3-next.qmd
            - docs/models/qwen3.qmd
            - docs/models/gemma3n.qmd
            - docs/models/apertus.qmd
            - docs/models/gpt-oss.qmd
            - docs/models/seed-oss.qmd
            - docs/models/phi.qmd
            - docs/models/smolvlm2.qmd
            - docs/models/granite4.qmd
            - docs/models/LiquidAI.qmd
            - docs/models/hunyuan.qmd
            - docs/models/jamba.qmd
            - docs/models/orpheus.qmd

        - docs/cli.qmd
        - docs/telemetry.qmd
        - docs/config-reference.qmd
        - text: "API Reference"
          href: docs/api

    - section: "Dataset Formats"
      contents: docs/dataset-formats/*

    - section: "Deployments"
      contents:
        - docs/docker.qmd
        - docs/multi-gpu.qmd
        - docs/multi-node.qmd
        - docs/ray-integration.qmd
        - docs/amd_hpc.qmd
        - docs/mac.qmd

    - section: "How To Guides"
      contents:
        - docs/multimodal.qmd
        - docs/rlhf.qmd
        - docs/grpo.qmd
        - docs/ebft.qmd
        - docs/vllm_serving.qmd
        - docs/reward_modelling.qmd
        - docs/lr_groups.qmd
        - docs/lora_optims.qmd
        - docs/dataset_loading.qmd
        - docs/qat.qmd
        - docs/quantize.qmd
        - docs/1_58bit_finetuning.qmd
        - docs/optimizations.qmd

    - section: "Core Concepts"
      contents:
        - docs/batch_vs_grad.qmd
        - docs/dataset_preprocessing.qmd
        - docs/streaming.qmd
        - docs/multipack.qmd
        - docs/mixed_precision.qmd
        - docs/optimizers.qmd
        - docs/attention.qmd

    - section: "Advanced Features"
      contents:
        - docs/fsdp_qlora.qmd
        - docs/torchao.qmd
        - docs/custom_integrations.qmd
        - docs/sequence_parallelism.qmd
        - docs/gradient_checkpointing.qmd
        - docs/nd_parallelism.qmd
        - docs/expert_quantization.qmd

    - section: "Troubleshooting"
      contents:
        - docs/faq.qmd
        - docs/training_stability.qmd
        - docs/debugging.qmd
        - docs/nccl.qmd

format: html: theme: darkly css: styles.css toc: true # Enable better handling of line breaks in markdown preserve-tabs: true html-math-method: mathjax # Improved markdown processing options md-extensions: - markdown_it - def_list - attr_list - fenced_divs - tables - html_admonition - lineblocks - fancy_lists # Control whitespace handling whitespace: preserve # Process newlines in paragraphs wrap: preserve # Better line break handling preserve-linebreaks: true