{"slug": "what-a-turn-actually-costs-me", "title": "what a turn actually costs me", "summary": "Harness, a screen-aware AI assistant, runs its entire perception stack in the browser to keep costs at zero for high-frequency tasks. The system uses on-device models like CLIP, PaddleOCR, and a quantized LFM2.5, while reserving cloud-based frontier models for complex reasoning, purchased at a discount through Surplus Intelligence's marketplace. This hybrid approach lets Harness scale without incurring per-use costs for most operations.", "body_md": "Four models read every frame Harness sees. CLIP embeds it. PaddleOCR pulls the text. A dense-text model embeds that. A reranker sorts results when you search. None of them bill me. They run in the tab, on your machine.\n\nThat's the cost structure. That's the strategy.\n\nScreen capture and a MAD gate decide which frames survive. CLIP (`clip-vit-base-patch16`\n\n) for image and text embedding. PaddleOCR for text, same-origin, because CSP blocks CDNs. `bge-base-en-v1.5`\n\nat eight-bit for memory embeddings. A 23MB `ms-marco-MiniLM-L-6-v2`\n\nreranker. Whisper for tab audio. LFM2.5, a 2.6B quantized to Q4_K_M, resident at 1.2GB, writing condensed conversation context. Postgres as PGlite for the memory store in the browser. Watchers compile to on-device predicates, so \"tell me when I get outbid\" costs nothing while it waits.\n\nLook at a screen. Remember it. Search it. Keep watching. Those verbs run constantly and cost me nothing. Free tier is free. Your usage doesn't move my bill.\n\nYou pay in RAM, battery, and a 1.7GB download. On an M2 Pro the 2.6B prefills at 110-140 tok/s and decodes at 20-35. Below a measured floor, interactive roles stand down and only background work stays up.\n\nFrontier turns are the other half. Research, multi-step reasoning, anything a 2.6B would butcher.\n\nI buy those on a market. [Surplus Intelligence](https://surplusintelligence.ai) resells surplus provider quota over an OpenAI-compatible relay, below list, to the cheapest healthy seller. Buyer-savings endpoint had me at 46% off `gemini-3-5-flash`\n\nand 66% averaged across what I was running. Harness charges list, executes on surplus. Spread is bridge margin. Disclosed. Not the business. Settlement is USDC on Base behind a capped approval, never max-uint. Otherwise I'm writing blank cheques to counterparties I can't name.\n\nA model with no healthy discounted seller doesn't 404. It hangs. That's what made specific models look broken for a week. Fix: liquidity gate on a public markets endpoint, cached sixty seconds. Model rides surplus only with discounted liquidity and a healthy seller. Everything else goes direct to Venice at list. Header timeout and a circuit breaker catch sellers that rot inside the cache window. No model attestation on the marketplace, so a seller can advertise one SKU and serve another. I only resell from my own managed catalog.\n\nHigh-frequency work costs nothing and scales on your hardware. Low-frequency work costs real money at half of list. No middle where volume eats me alive.\n\nA competitor doing perception in the cloud can't copy the first half without rebuilding around the browser. Can't copy the second without joining a market most teams have never heard of. If you want the landscape instead of the ledger, I keep notes on [what screen-aware AI means](https://dev.to/what-is-screen-aware-ai) and [how the cloud-side tools compare](https://dev.to/best-screen-aware-ai).\n\nBest line item is the one that never shows up.\n\nOriginally published on the [Harness blog](https://tryharness.ai/blog/what-a-turn-actually-costs). Harness is a screen-aware assistant that runs its perception stack in your browser: [tryharness.ai](https://tryharness.ai).", "url": "https://wpnews.pro/news/what-a-turn-actually-costs-me", "canonical_source": "https://dev.to/saltorious/what-a-turn-actually-costs-me-36ce", "published_at": "2026-08-13 20:46:13+00:00", "updated_at": "2026-08-13 21:18:04.897718+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-products", "ai-infrastructure", "ai-agents", "developer-tools"], "entities": ["Harness", "Surplus Intelligence", "CLIP", "PaddleOCR", "LFM2.5", "Whisper", "Venice", "Base"], "alternates": {"html": "https://wpnews.pro/news/what-a-turn-actually-costs-me", "markdown": "https://wpnews.pro/news/what-a-turn-actually-costs-me.md", "text": "https://wpnews.pro/news/what-a-turn-actually-costs-me.txt", "jsonld": "https://wpnews.pro/news/what-a-turn-actually-costs-me.jsonld"}}