ByteDance · 2.7B parameters · 65.5K context · commercial OK
Local (your GPU) $0 · ~10 GB VRAM | Rent a GPU from $0.06/hr | Your API key $0.12/1M est. |
Watch this price — free
Get an email when Ouro 2.6B's tracked price drops — checked against the append-only daily record. No account needed.
Pro watches every price with exact targets, by email or signed webhook — see pricing.
Ouro 2.6B — 2.7B params (ByteDance).
At $0.12/1M tokens, the cheapest API for Ouro 2.6B stays cheaper than a used RTX 3090 at any daily volume — electricity alone costs more per token than the API does.
Local is never $0 — every figure here includes electricity and hardware amortization. Speeds come from published 7-8B Q4 benchmarks (your model, quant, and context will differ); break-even volumes are rounded to 2 significant figures.
| Parameters | 2.7B |
|---|---|
| Context window | 65.5K tokens |
| Recommended quant | Q4_K_M |
| VRAM requirement | ~10 GB (at Q4_K_M, 16.4K context) |
| Download size | ~2 GB |
| License | Commercial use OK |
Open the free Spanvero advisor → for the live, interactive math for your exact workload and hardware.
Three ways: $0 on your own machine if you have about 10 GB of VRAM (at Q4_K_M); a rented GPU from $0.06/hr (NVIDIA RTX 3060 12GB at the direct Vast.ai price); or about $0.12 per 1M blended tokens via your own API key (a rough estimate for this size). Spanvero adds $0 markup on every path.
Yes — Ouro 2.6B runs locally for $0 in compute if your GPU or Apple Silicon Mac has about 10 GB of VRAM/unified memory (at Q4_K_M with 16.4K context). Point LM Studio, Ollama, or llama.cpp at it; after the download it runs offline and free.
Any GPU or Apple Silicon Mac with about 10 GB of VRAM/unified memory handles Ouro 2.6B at Q4_K_M. Of the consumer hardware Spanvero tracks, the smallest that fits is a used NVIDIA RTX 3080 10GB.
Browse: All models · Compare
Spanvero · What's new · Prices as of 2026-07-13. We're an honest advisor — $0 markup, your own accounts, we never resell compute. Catalog specs auto-indexed from the Hugging Face Hub — parameter count is exact; download size and quantizations are estimates. © 2026 Cynosure LLC.
A short email of real AI price moves, straight from the daily log — no hype. We're collecting the list now; the first issue goes out when it opens. Unsubscribe with one click.