Results for “compression”

Context compression for YAMTAM — nén JSON/structured tool output trước khi vào LLM. Hiệu quả với JSON (50-72% tiết kiệm); text thuần cần bản [all].

Extends effective context capacity through strategic compression, masking, caching, and partitioning techniques.

Decides when to delegate to compressed-output subagents (investigator, builder, reviewer) instead of using vanilla agents, saving main-context tokens across long sessions.

Enhances AI agent memory by injecting large external context, auto-summarizing conversations with tone and intent preservation, compressing multi-turn histories, and verifying facts.

Provides structured guidance for building production-grade AI agent systems through context engineering, covering fundamentals, architectural patterns, operational excellence, and evaluation.

Enables compact, machine-native agent-to-agent messaging using a shared vocabulary of 340+ 2-character atoms across 7 domains, achieving 3x compression over natural language.

Quantize large language models to 8/4/3/2/1-bit precision without calibration data, using multiple optimized backends for deployment with vLLM or HuggingFace Transformers.

Enables persistent ultra-compressed technical communication; use for explicit brevity or token-reduction requests, except where fragments risk safety or clarity.

Injects up to 1 million tokens of external context into AI agent memory, auto-summarizes conversations with tone and intent preservation, and compresses multi-turn history into dense token sequences.

Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.

Compress large language models using knowledge distillation from teacher to student models, covering temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.

Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.

Compress large language models by 40-60% with minimal accuracy loss using one-shot pruning techniques like Wanda and SparseGPT, enabling faster inference and deployment on constrained hardware.

Shares memory between agents at the representation level by compacting the orchestrator's KV cache for efficient worker handoff, reducing token costs without summarization or retrieval.

Use the ClawHub CLI to search, install, update, and publish agent skills from clawhub.ai with advanced caching and compression. Use when you need to fetch new skills on the fly, sync installed skills to latest or a specific version, or publish new/updated skill folders with optimized performance.

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

TIGER v2 — Multi-scanner trading system for Hyperliquid perps via Senpi MCP. 5 signal patterns (BB compression breakout, BTC correlation lag, momentum breakout, mean reversion, funding rate arb), DSL v4 trailing stops, goal-based aggression engine, and risk guardrails. Configurable profit target over deadline. 12-cron architecture (10 TIGER + prescreener + ROAR meta-optimizer). Pure Python analysis. Requires Senpi MCP, python3, mcporter CLI, and OpenClaw cron system.

Use when asked to set up AI agent templates with coordinator mode, implement persistent agent memory, compress context for long-running agents, build multi-agent workflows with skills/agents/workflows structure, or use the .agents/ folder convention for AI-native editors. Triggers on: 'ag-kit', 'antigravity kit', 'coordinator mode agent', 'persistent agent memory', 'context compression agent', 'agent workflow template', 'multi-agent coordinator', '.agents folder', 'agent skills workflows', 'ag kit init', 'template đa agent', 'bộ nhớ agent lâu dài'.