Full fine-tuning recipe: QLoRA (4-bit) on Mistral Small 3 via Axolotl, targeting 2x RTX 4090, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Mistral Nemo 12B via LitGPT, targeting AWS g5.12xlarge, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Qwen 2.5 32B via torchtune, targeting AWS g5.12xlarge, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Qwen 2.5-Coder 7B via Unsloth, targeting AWS g5.12xlarge, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Gemma 2 9B via OpenRLHF, targeting AWS p4d.24xlarge, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Phi-3.5-mini via torchtune, targeting AWS p4d.24xlarge, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Phi-4 via Unsloth, targeting Lambda Labs 8xH100, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Mixtral 8x7B via OpenRLHF, targeting Lambda Labs 8xH100, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Mixtral 8x22B via DeepSpeed, targeting single A100 80GB, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Llama 3.3 70B via Hugging Face TRL, targeting single A100 80GB, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Llama 3.1 8B via Megatron-LM, targeting single A100 80GB, with data mix and eval plan.
Full fine-tuning recipe: QLoRA (4-bit) on Mistral Small 3 via FSDP, targeting single H100 80GB, with data mix and eval plan.