hiyouga/LlamaFactory
Documented errors, page 3 of 4. Back to hiyouga/LlamaFactory
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| KTransformers inference requires `infer_backend… | validation | error | ktransformers, vllm, sglang, inference, llamafactory |
| RM training dataset is empty | validation | error | data, dataset, reward-model, validation |
| SGLang engine does not support `get_scores`. | exception | error | sglang, not-implemented, reward-model, inference |
| Batch infer is not implemented. | exception | error | inference, not-implemented, v1-api |
| File access is restricted to the safe media directory. | http | error | security, lfi, path-traversal, http-403, multimodal |
| KTransformers does not support lora reward model. | validation | error | ppo, ktransformers, reward-model, lora, config |
| Local file access is disabled. | http | error | security, lfi, api, http-403, multimodal |
| Turn off `streaming` when saving dataset to disk. | exception | error | streaming, tokenized-path, config, dataset |
| Unknown mixing strategy | exception | error | dataset, mixing, config, validation |
| vLLM engine does not support bnb quantization (GPTQ and AWQ… | validation | error | vllm, bnb, quantization, gptq, awq |
| Buffer size exceeds max buffer size | exception | error | buffer, memory, generation, validation |
| DeepSpeed config_file is required in dist_config | validation | error | deepspeed, config, distributed, zero3 |
| Distributed training does not support layer-wise GaLore. | validation | error | galore, optimizer, distributed, layerwise, incompatible-flags |
| `kt_model_max_length` must be a positive integer. | validation | error | ktransformers, config, validation, llamafactory |
| `predict_with_generate` is not supported in KTransformers… | exception | error | ktransformers, sft, evaluation, generation |
| vLLM/SGLang backend is only available for API, CLI and Web. | validation | error | vllm, sglang, infer-backend, training, config-validation |
| The installed Transformers-KT does not provide… | exception | critical | ktransformers, dependencies, import-error, llamafactory |
| The model is not a `PreTrainedModel`, export aborted. | exception | error | export, ktransformers, model-type |
| Quantization is only compatible with the LoRA or OFT method. | validation | error | quantization, qlora, config, llamafactory |
| Stop words are required to replace the EOS token. | exception | error | template, tokenizer, custom-template |
| Dataset converter not found. | exception | error | dataset, converter, registry, config |
| No checkpoint files found in | validation | error | fsdp2, model-loading, checkpoint, filesystem |
| No JSON/JSONL files found in the specified path | exception | error | cloud-storage, dataset, file-format, fsspec |
| Number of videos ( ) must match number of audios ( ) when… | exception | error | multimodal, qwen2-omni, audio, video, dataset-validation |
| Please set adapter_name_or_path to merge adapters into base… | validation | error | peft, export, lora, configuration |
| Template is required for MultiModalDataCollator. | exception | error | collator, multimodal, config, training |
| A placeholder is required in the string formatter. | exception | error | formatter, template, config, custom-template |
| FLA operator ` ` did not match any model module attributes. | exception | error | runtime, model-mismatch, kernel-plugin, version-skew |
| hyper_parallel is not installed. Please install it with… | exception | error | hyper-parallel, fsdp, optional-dependency, installation |
| Cannot build a dummy media fragment for a text-only… | exception | error | multimodal, text-only, api-misuse, processor |
| mp_replicate_size * mp_shard_size must equal to world_size… | exception | error | distributed-training, v1, configuration, device-mesh |
| Please enable `predict_with_generate` to save model… | validation | error | config, validation, prediction, sft, generation |
| ep_size must be positive, got | validation | error | fsdpturbo, expert-parallel, distributed, configuration |
| KTransformers requires a local adapter directory. | validation | error | ktransformers, adapter, filesystem, llamafactory |
| `save_epochs` is not supported with `dynamic_batching`; use… | exception | error | v1, configuration, checkpointing, batching |
| Distributed training does not support layer-wise APOLLO. | validation | error | apollo, optimizer, distributed, layerwise, incompatible-flags |
| `kt_cpu_activation` is only valid when `use_kt: true`. | validation | error | ktransformers, config-validation, dependent-fields |
| `save_dir` already exists, use another one. | validation | error | config, filesystem, evaluation, idempotency |
| `tensor_model_parallel_size` must be >= 1. | validation | error | config, megatron, parallelism, distributed |
| cuda_fused_moe requires Triton. | exception | error | dependencies, triton, cuda, moe |
| CudaFusedMoEKernel requires CUDA, current accelerator is | exception | error | device-check, cuda, moe, kernel-plugin |
| MOSS-VL media tokens do not match the provided media after… | exception | error | multimodal, moss-vl, truncation, cutoff-len, media-count |
| `neat_packing` cannot be set as True except SFT. | validation | error | config, validation, packing, stage, sft |
| Only HTTP/HTTPS URLs are allowed. | http | error | security, ssrf, url-validation, http-400 |
| MOSS-VL media lengths do not consume all provided inputs. | exception | error | multimodal, moss-vl, batching, truncation, media-count |
| PPO training is incompatible with S^2-Attn. | validation | error | config, ppo, attention, long-context, validation |
| Unsupported quantization bit | validation | error | quantization, bitsandbytes, configuration |
| No valid messages or chosen_messages/rejected_messages… | validation | error | data-format, dataset, preference-data, rendering |
| This device does not support `pure_bf16`. | validation | error | pure-bf16, hardware, gpu-compat, precision, config-validation |
| `use_dora` is only valid for LoRA training. | validation | error | config, lora, dora |
| `adapter_folder` must stay inside the KT adapter directory. | validation | error | ktransformers, adapter, path-traversal, config |
| KTransformers supplies its checkpoint context; remove… | validation | error | ktransformers, gradient-checkpointing, config, llamafactory |
| dim must be specified. | exception | error | v1, api-misuse, distributed-training, device-mesh |
| Disable FSDP activation checkpointing when using… | validation | error | ktransformers, fsdp, config, llamafactory |
| Fewer tags than provided media files ( unused). | validation | error | data, multimodal, dataset-conversion, validation |
| Path does not exist: . | exception | error | cloud-storage, fsspec, dataset, permissions |
| Unsloth does not support lora reward model. | validation | error | ppo, unsloth, reward-model, lora, config |
| vLLM engine does not support `get_scores`. | exception | error | vllm, not-implemented, reward-model, inference |
| Cannot open due to . | exception | error | dataset-config, json, file-io, hf-hub, config-path |
| `predict_with_generate` is incompatible with DeepSpeed… | validation | error | deepspeed, zero-3, predict-with-generate, evaluation, incompatible-flags |
| Template already exists. | exception | error | template, custom-template, idempotency |
| compute_dtype must be str or torch.dtype, got | validation | error | quantization, bitsandbytes, dtype, type-error |
| HQQ only accepts 1/2/3/4/5/6/8-bit quantization. | exception | error | hqq, quantization, config-validation |
| vLLM only accepts a single adapter. Merge them first. | validation | error | vllm, lora, adapter-merge, export |
| DoRA is not compatible with PTQ-quantized models. | validation | error | dora, lora, gptq, awq, quantization |
| Please use scripts/pissa_init.py to initialize PiSSA in… | validation | error | pissa, lora, deepspeed, zero-3, incompatible-flags |
| world_size ( ) must be divisible by cp_size ( ). | exception | error | distributed-training, v1, context-parallelism, configuration |
| Cannot specify `val_size` if `eval_dataset` is not None. | exception | error | config, data-args, validation-split |
| `pissa_init` is only valid for LoRA training. | validation | error | config, lora, pissa |
| PPO training does not support evaluation, use the SFT stage… | validation | error | config, ppo, validation, evaluation |
| Quantized models are not supported with Megatron Bridge. | validation | error | megatron, quantization, config, validation |
| Each must be followed by an when using audio in video. | exception | error | multimodal, qwen2-omni, audio, video, placeholder-order |
| DeepSpeed only supports bf16 mixed precision for now, fp16… | validation | error | deepspeed, precision, config, distributed |
| kernel_config.include_kernels must select at least one FLA… | validation | error | configuration, validation, kernel-plugin |
| Megatron Bridge only supports the `pt` and `sft` stages. | validation | error | megatron, config, stage, validation |
| This model does not support video input. Please check… | exception | error | multimodal, video, template, config |
| vLLM engine does not support RoPE scaling. | validation | error | vllm, rope-scaling, long-context, infer-config |
| Please merge adapters before quantizing the model. | exception | error | export, quantization, lora, merge |
| Unsupported `flash_attn | exception | error | v1, configuration, attention, enum-validation |
| `use_rslora` is only valid for LoRA training. | validation | error | config, lora, rslora |
| Cannot get scores using an auto-regressive model. | exception | error | model-loading, hf-engine, scoring, reward-model |
| Megatron Bridge is incompatible with DeepSpeed. | validation | error | megatron, deepspeed, distributed, config, backend-conflict |
| This model does not support audio input. Please check… | exception | error | multimodal, audio, template, config |
| File types should be identical. | exception | error | dataset, file-format, directory, config |
| MOSS-VL batch metadata must have one entry per sample. | exception | error | multimodal, moss-vl, batching, internal-invariant |
| Multimodal plugin already exists. | exception | error | plugin-registry, multimodal, extension-api, name-collision |
| Sequence parallelism requires flash attention. Please set… | exception | error | v1, context-parallelism, flash-attention, configuration |
| Can not load dataset from | validation | error | data, path, dataset-loading, config |
| KTransformers thin integration currently supports LoRA… | validation | error | ktransformers, lora, config, llamafactory |
| has no RMSNorm weight for NPU RMSNorm kernel. | exception | error | runtime, npu, rmsnorm, model-state |
| requires a gate tensor for NPU Gated RMSNorm. | validation | error | runtime, npu, rmsnorm, api-contract |
| MOSS-VL encountered a video end token without a matching… | exception | error | multimodal, moss-vl, video, truncation, cutoff-len |
| Plugin configuration must have a 'name' field. | exception | error | v1, plugins, configuration, validation |
| Please make sure eval_dataset be provided or val_size >1e-6 | validation | warning | evaluation, dataset, val-size, config-validation |
| DeepSpeed config_file is required. | validation | error | config, deepspeed, distributed, validation |
| Flash Linear Attention and FSDPTurbo are required for this… | exception | error | kernels, fla, fsdp-turbo, dependencies, install |
| dp_size must be divisible by ep_size, got | validation | error | fsdpturbo, expert-parallel, distributed, topology |
| KTransformers accepts a single `adapter_name_or_path`. | validation | error | ktransformers, lora, adapter, llamafactory |
| The current model does not support `stream_chat`. | exception | error | model-loading, hf-engine, streaming, model-type |
| The length of eval dataset and interleave probs should be… | validation | error | config, data-args, mixing |