DeepTutor
Open source, agent native learning workspace for personalized tutoring, quizzes, research and practice
ScaleDown
Task-specific models for classify, extract, summarize, compress
About
Task-specific NLP models for production pipelines
ScaleDown ships four task-specific models (Classify, Extract, Summarize, Compress) with a 1M-token context window. Define labels or entities in plain English; no fine-tuning required. Pricing is flat at $0.05 per 1M input tokens (image input $0.15/1M); output tokens are free. New accounts get 1M free tokens. Aimed at tagging, triage, structured extraction, summarization at scale, and query-aware context compression before sending work to a frontier model. SOC 2, HIPAA, GDPR, ISO 27001 posture; zero data retention and self-hosted options.
Features
- Classify, Extract, Summarize, Compress models
- 1M token context, no fine-tuning required
- Flat $0.05/1M input tokens; free output
- 1M free tokens on signup
- Self-hosted / dedicated enterprise options
- SOC 2, HIPAA, GDPR, ISO 27001
Related skills
All skillsRelated articles
All articlesTavily vs Parallel (2026): Web Search APIs for AI Agents
I compare Tavily and Parallel as web search APIs for AI agents: free tiers, credit and per-request pricing, MCP servers and open source status.
Langfuse vs Braintrust (2026): Pricing, Self-Hosting, MCP
I compare Langfuse and Braintrust for LLM tracing and evals: free tiers, paid plans, self-hosting, open source status and their MCP servers.
5 Best Web Search and Browser Tools for AI Agents (2026)
The APIs and cloud browsers that let an agent search, read and use the live web, compared on what they return, how they bill and how to plug them in.
More Research
View all
Hotelist MCP
Free, no signup MCP server for hotel search, with AI ratings from reviews and photos, by @levelsio
Kolect
Find and vet TikTok, Instagram and YouTube creators in one sentence, inside Claude or Codex