Customer supportLoRALlama 3.1 8B Instruct
Fine-tune a support-ticket classifier
Route and tag customer support tickets with a private Llama-3.1-8B LoRA — trained in minutes, served for cents.
Llama 3.1 8B Instruct~12 min~$0.30 / M tokens
Showcase · 20 ready-to-ship builds
See what you can build in minutes.
A gallery of private models you can fine-tune and serve today — classifiers, copilots, and brand-voice writers. Start one with a single click, or open the sandbox and feel it first.
Developer toolsQLoRAQwen3 8B
Build a private SQL copilot
Turn natural language into your schema's SQL with a Qwen3-8B QLoRA — your queries never leave your tenant.
Qwen3 8B~15 min~$0.30 / M tokens
MarketingLoRAMistral Small 3.1 24B
Train a model in your brand voice
Fine-tune a private writer that sounds exactly like your brand — on your own copy, never shared.
Mistral Small 3.1 24B~20 min~$0.45 / M tokens
Developer toolsLoRAQwen3 8B
Fine-tune a code-review assistant
A private reviewer trained on your codebase's conventions — flags issues in your house style, in your tenant.
Qwen3 8B~14 min~$0.30 / M tokens
Knowledge workLoRALlama 3.1 8B Instruct
Train a domain document summarizer
Summarize contracts, papers, or reports in your domain's style with a private Llama-3.1-8B adapter.
Llama 3.1 8B Instruct~12 min~$0.30 / M tokens
Conversational AIFast LoRAGemma 3 12B IT
Fast intent classifier on Gemma 3 12B
Mint an intent classifier in minutes with fast fine-tuning on Gemma 3 12B.
Gemma 3 12B IT~4 min~$0.35 / M tokens
Retrieval / RAGLoRAQwen3 8B
Fine-tune a RAG reranker
Reorder retrieved passages by true relevance to your domain — a private Qwen3-8B LoRA that lifts answer quality without touching your retriever.
Qwen3 8B~14 min~$0.30 / M tokens
Agents / toolsLoRALlama 3.1 8B Instruct
Train a reliable function-calling model
Teach a private Llama-3.1-8B to emit valid tool calls for YOUR API surface — fewer malformed calls, no schema drift.
Llama 3.1 8B Instruct~16 min~$0.30 / M tokens
Customer supportLoRAMistral Small 3.1 24B
Fine-tune an email triage + responder
Classify inbound email and draft on-policy replies with a private Mistral-Small adapter — your inbox never leaves your tenant.
Mistral Small 3.1 24B~22 min~$0.45 / M tokens
Legal / knowledge workLoRALlama 3.1 8B Instruct
Extract contract clauses to JSON
Pull parties, terms, dates and obligations out of contracts into structured JSON — a private Llama-3.1-8B LoRA for your paper.
Llama 3.1 8B Instruct~15 min~$0.30 / M tokens
E-commerceFast LoRAGemma 3 12B IT
Fast e-commerce product categorizer
Map messy product titles to your taxonomy in minutes with fast fine-tuning on Gemma 3 12B.
Gemma 3 12B IT~5 min~$0.35 / M tokens
AnalyticsLoRALlama 3.1 8B Instruct
Aspect-based sentiment analysis
Score sentiment per aspect (price, quality, service…) from reviews with a private Llama-3.1-8B LoRA.
Llama 3.1 8B Instruct~12 min~$0.30 / M tokens
Conversational AILoRAMistral Small 3.1 24B
Build a custom-persona chatbot
A chat model that stays in character and on-policy — fine-tune a private Mistral-Small persona on your transcripts.
Mistral Small 3.1 24B~21 min~$0.45 / M tokens
Developer toolsLoRAQwen3 8B
Generate docstrings in your style
Auto-write docstrings that match your repo's conventions with a private Qwen3-8B LoRA.
Qwen3 8B~14 min~$0.30 / M tokens
Information extractionLoRALlama 3.1 8B Instruct
Custom named-entity extractor
Extract your domain's entities (SKUs, gene names, ticket IDs…) with a private Llama-3.1-8B LoRA — no off-the-shelf NER fits your schema.
Llama 3.1 8B Instruct~13 min~$0.30 / M tokens
LocalizationLoRAQwen3 8B
Domain-specialized translation
Translate in your domain's terminology and register with a private Qwen3-8B LoRA — generic MT gets your jargon wrong.
Qwen3 8B~16 min~$0.30 / M tokens
ProductivityLoRALlama 3.1 8B Instruct
Meeting transcript → action items
Turn raw transcripts into owners, decisions and action items with a private Llama-3.1-8B LoRA.
Llama 3.1 8B Instruct~14 min~$0.30 / M tokens
Trust & safetyFast LoRAGemma 3 12B IT
Fast content moderation classifier
Flag policy violations against YOUR rules in minutes with fast fine-tuning on Gemma 3 12B.
Gemma 3 12B IT~5 min~$0.35 / M tokens
HR techLoRAQwen3 8B
Parse resumes to structured JSON
Extract skills, roles and dates from resumes into clean JSON with a private Qwen3-8B LoRA.
Qwen3 8B~13 min~$0.30 / M tokens
Model alignmentDPOLlama 3.1 8B Instruct
Align a model with DPO preference data
Make your assistant prefer your good answers over your bad ones with Direct Preference Optimization on Llama-3.1-8B.
Llama 3.1 8B Instruct~25 min~$0.30 / M tokens
More builds on the way
We are adding recipes for new tasks and base models. Have one you want next? Spin up your own in the dashboard — any HF model works.
Start from scratch Sandbox
Prefer to play first?
Open the Playground and prompt a live model in a safe sandbox — no setup, no commitment. Tune the inputs, watch the output, then turn what you like into a real fine-tune.
No setup required Isolated sandbox