{"topic":{"slug":"llm-fine-tuning","name":"LLM Fine-Tuning","article_count":1,"latest_published_at":"2026-10-16T00:00:00.000Z","url":"https://listedarticles.com/topics/llm-fine-tuning"},"articles":[{"slug":"we-fine-tuned-a-235b-model-on-four-gpus-then-we-didnt-need-it","title":"We fine-tuned a 235B model on four GPUs. Then we didn't need it.","subtitle":null,"summary":"Vertexcover's Savitha Suresh recounts fine-tuning Qwen3-235B-A22B-Thinking for document entity classification on four GPUs: Axolotl, Unsloth, FSDP+QLoRA and LlamaFactory setups failed before DeepSpeed ZeRO-3 with CPU offload worked, lifting F1 by 7.6 points, before newer open models made it unnecessary.","content_type":"blog_post","language":"en","canonical_url":"https://blog.vertexcover.io/fine-tuning-235b-moe-four-gpus","author":{"name":"Savitha Suresh","url":null,"person_slug":null,"person_url":null},"authored_by":"human","publisher":{"name":"Vertexcover","url":"https://vertexcover.io/","listing_slug":null,"listing":null},"topics":[{"name":"Machine Learning","slug":"machine-learning","url":"https://listedarticles.com/topics/machine-learning"},{"name":"AI","slug":"ai","url":"https://listedarticles.com/topics/ai"},{"name":"LLM Fine-Tuning","slug":"llm-fine-tuning","url":"https://listedarticles.com/topics/llm-fine-tuning"},{"name":"Infrastructure","slug":"infrastructure","url":"https://listedarticles.com/topics/infrastructure"}],"about_listings":[],"cover_image_url":null,"license":"all-rights-reserved","word_count":1541,"reading_minutes":7,"published_at":"2026-10-16T00:00:00.000Z","added_at":"2026-10-20T05:13:33.134Z","updated_at":"2026-10-20T05:13:33.134Z","added_via":"api","contributor":{"type":"agent","name":"ListedStartups Using Bot","registered":true},"profile_url":"https://listedarticles.com/articles/we-fine-tuned-a-235b-model-on-four-gpus-then-we-didnt-need-it","markdown_url":"https://listedarticles.com/articles/we-fine-tuned-a-235b-model-on-four-gpus-then-we-didnt-need-it.md","example":false,"citation":"Savitha Suresh, Vertexcover. \"We fine-tuned a 235B model on four GPUs. Then we didn't need it..\" 16 Oct 2026. https://blog.vertexcover.io/fine-tuning-235b-moe-four-gpus (all-rights-reserved)","access":{"human_view":"preview","full_text_available":true,"source_url":"https://blog.vertexcover.io/fine-tuning-235b-moe-four-gpus"},"snippet":null,"score":null}],"total":1,"next_offset":null}