Rebiha

Home › Datasets › Cooking & Recipes

Cooking & Recipes Fine-Tuning Dataset

Recipes, techniques, substitutions, and kitchen guidance.

Fine-tune a model on this dataset

$19 for the dataset. Training is priced separately by model, starting at $47.

What is in this dataset

902 instruction-response pairs covering recipes, techniques, substitutions, and kitchen guidance. Each pair is a realistic question with a full written answer, in the JSONL format used for supervised fine-tuning. No synthetic filler, no duplicated rows.

Example pairs

Real rows from the dataset, shown in full.

I need a quick weeknight dinner recipe that takes under 30 minutes. I have chicken breast, garlic, cherry tomatoes, and pasta. What can I make?

Make a 20-minute Garlic Tomato Chicken Pasta. Start by bringing a large pot of salted water to a boil and cooking 8 oz of pasta (penne or spaghetti) according to package directions. Meanwhile, slice 2 boneless, skinless chicken breasts in half horizontally to create thin cutlets, season with salt and pepper. Heat 2 tbsp olive oil in a large skillet over medium-high heat. Sear the chicken cutlets for 3 minutes per side until golden and cooked through (165°F internal). Remove and set aside. In the same skillet, reduce heat to medium, add 3 minced garlic cloves and sauté for 30 seconds. Add 2 cups halved cherry tomatoes, 1/4 tsp red pepper flakes, and a pinch of salt. Cook for 3–4 minutes…

Can you explain the proper technique for searing meat? I always end up with a gray, steamed crust instead of a brown crust.

Searing is about high heat and dry surfaces. First, pat your meat completely dry with paper towels—moisture is the enemy of browning. Season generously with salt and pepper just before searing, not ahead of time (salt draws out moisture if left too long). Choose a heavy pan like cast iron or stainless steel; non-stick won't get hot enough for a good crust. Heat the pan over medium-high to high heat for 2–3 minutes until it's smoking hot. Add a high smoke-point oil like avocado or canola—just enough to coat the bottom (about 1 tbsp). Carefully place the meat in the pan, laying it away from you to avoid oil splatter. Do not move it for at least 3–4 minutes. The meat will release…

I'm making a vinaigrette but it keeps separating immediately. What's the proper emulsifying technique?

Emulsifying a vinaigrette requires a stable suspension of oil and acid. Start with a ratio of 3 parts oil to 1 part acid (e.g., 3 tbsp olive oil to 1 tbsp vinegar). Use a bowl with a rounded bottom and a whisk, or a jar with a tight lid. First, whisk together your acid (vinegar or lemon juice), a pinch of salt, pepper, and any flavorings (mustard, honey, minced shallot). Dijon mustard (about 1 tsp per 3 tbsp oil) is a natural emulsifier and helps stabilize the mix. Then, slowly drizzle the oil in a thin, steady stream while whisking vigorously—this physically breaks the oil into tiny droplets suspended in the acid. If you pour too fast, the emulsion breaks. For a more stable vinaigrette,…

Who this is for

Recipe apps and kitchen assistants covering technique and substitutions.

Which models this works with

The dataset is plain JSONL, so it works with any instruction-tuned open model. On Rebiha you can train it directly on:

Training uses LoRA, or QLoRA on a 4-bit base for larger models. Your base model's weights are never modified — training produces a separate adapter.

How it works

  1. Pick this dataset and a base model.
  2. Rebiha runs the job on a managed GPU. You set up nothing.
  3. Download your model: a ready-to-run GGUF, plus developer assets (adapter, tokenizer, configs).

Fine-tuning or RAG?

Use RAG when the model needs facts that change — your prices, your documents, today's inventory. Use fine-tuning when you want consistent style and structure without prompting for it every time. This dataset teaches the shape of a good answer in this domain, not facts to look up.

Related datasets

Start fine-tuning