// AI TOOLS · BONSAI 27B

Bonsai 27B

PrismML's compressed 27B multimodal model — quantized to 1-bit and ternary weights so a model class that used to live in the cloud runs fully on-device, including on a phone.

KompozyTurn one idea into a week of content — across every platform, published for you.
Get Started →

Last verified · 2026-07-14 · by Moe Ameen

What Bonsai 27B is

Bonsai 27B is a compressed large language model from PrismML, a US-based AI company that emerged from a team of Caltech researchers with backing from Khosla Ventures, Cerberus, Google, and Samsung. It launched on July 14, 2026, and the headline claim is that it is the first 27B-class model that runs on a phone — a size of model that until now effectively belonged in the cloud or on a workstation.

The trick is aggressive low-bit-width quantization. Bonsai is built on the open Qwen3.6 27B base and shipped in two variants: a ternary version that stores weights as {−1, 0, +1} (about a 5.9 GB footprint) and a 1-bit version that stores weights as binary {−1, +1} (about a 3.9 GB footprint). PrismML reports the ternary variant retains roughly 95% of the full-precision baseline's quality and the 1-bit variant about 90% — so the compression trades a modest amount of accuracy for a very small memory footprint. The 1-bit build is the one that fits within an iPhone 17 Pro's per-app memory budget.

It is multimodal — it accepts images as well as text via a compact 4-bit vision tower — and it carries a long 262K-token context window. Reported speeds range widely by hardware: on a phone the 1-bit variant runs at roughly 11 tokens/sec on an iPhone 17 Pro, while on desktop-class hardware it reaches far higher throughput (PrismML cites up to ~163 tokens/sec on an NVIDIA RTX 5090 and ~87 tokens/sec on an M5 Max for the 1-bit build). Everything is released under the Apache 2.0 license, with the GGUF weights on Hugging Face and a free, limited-time developer-preview API. Because it is a fast-moving first-of-its-kind release, confirm exact specs and any pricing on PrismML's own model card before you build against it.

What you can make with it

  • On-device text drafts — captions, hooks, thread outlines, short scripts, blog and newsletter starts — generated locally on a phone or laptop with no per-token API bill
  • Private brainstorming and reasoning that never leaves the device, useful for unreleased ideas or sensitive client work
  • Multimodal reads — point the model at a screenshot, a document, or camera input and get a text description, critique, or caption idea back
  • Agentic and tool-use loops (planning, refactoring, structured tool calls) for developers building local, offline-capable apps
  • Long-context analysis over a transcript or document pasted into the 262K window, done on-device
  • A capable offline assistant for drafting while traveling or filming, where connectivity is unreliable

How Kompozy turns Bonsai 27B output into content

The novel thing about Bonsai 27B is *where* it runs: in your pocket, offline, on the phone you already film with. That unlocks a specific workflow — capture and draft in the field, then finish and publish back at base. Say you shoot a walkthrough on location with no signal; Bonsai, running locally on the phone, can turn your rough voice notes into three caption angles and a hook while you're still standing there, no data connection needed. What it can't do is become the post. It writes no video, renders no image, sizes nothing for a feed, keeps no consistent brand voice across a week, and publishes nowhere. That's the exact handoff Kompozy is built for. Bring your on-device drafts and your footage into Kompozy, and its Persona Brief rewrites the raw text into your actual brand voice, then fans it into finished formats a phone model can't touch — Clipped Shorts from the footage with burned-in captions, a brand-exact Carousel through HyperFrames, Quote Graphics, Persona and HeyGen avatar video, plus native Text Posts, a Blog Article, and an Email Newsletter.

The pairing plays to Bonsai's strengths — mobility, privacy, and multimodal input. Snap a photo of a whiteboard or a competitor's post and let Bonsai read it and suggest a response on the spot; then Kompozy takes that conclusion and renders it into scheduled, on-brand content across nine social platforms plus email and blog, with Autopilot and a per-post review pipeline. And because Kompozy supports bring-your-own-key on the Founding tier, a team that leans on Bonsai for free, private, on-device drafting keeps that habit while Kompozy owns the media rendering, brand governance, and publishing the little model was never meant to do. Bonsai drafts anywhere, even with the plane in airplane mode; Kompozy turns those drafts into a finished week the moment you're back online.

  1. Run Bonsai 27B locally on your phone or laptop (1-bit for phones, ternary for more headroom) and draft your hooks, captions, and outlines on the spot — offline, private, and free per token. Paste in a transcript or point it at a screenshot to mine ideas.
  2. Bring the drafted text and any footage into Kompozy, and let the Persona Brief plus banned-word filters rewrite it into your consistent brand voice.
  3. Fan one idea into formats Bonsai can't make — a Carousel via HyperFrames, Quote Graphics, Persona or HeyGen avatar video, a blog recap, a newsletter, and Clipped Shorts from your video.
  4. Let Kompozy reframe each output per platform (9:16, 1:1, 16:9) and burn in branded captions.
  5. Schedule and publish the whole set across TikTok, Reels, Shorts, X, LinkedIn, and more from one queue with Autopilot.

Frequently asked questions

What is Bonsai 27B?

Bonsai 27B is a compressed large language model from PrismML, launched July 14, 2026. Built on the open Qwen3.6 27B base and quantized to 1-bit or ternary weights, it is described as the first 27B-class model to run on a phone. It is multimodal (reads text and images), has a 262K-token context, and is released under the Apache 2.0 license.

Can Bonsai 27B really run on a phone?

Yes. The 1-bit variant has roughly a 3.9 GB footprint, small enough to fit within an iPhone 17 Pro's per-app memory budget, where PrismML reports around 11 tokens/sec. The ternary variant (about 5.9 GB) needs more headroom but retains higher quality (~95% of the full-precision baseline vs ~90% for 1-bit). On desktop GPUs it runs far faster.

Is Bonsai 27B free to use?

The weights are free under the Apache 2.0 license — you can download the GGUF files from Hugging Face, run them locally, and use the model commercially. PrismML also offered a free, limited-time developer-preview API at launch. Your real cost is the device you run it on.

Can Bonsai 27B create and publish social media content?

No. It generates text and reads images on-device, but it renders no video, images, or designs, enforces no brand voice, and publishes to no platform. To turn its drafts into finished, on-brand posts across platforms you pair it with a content engine like Kompozy that renders the media and handles scheduling and publishing.

Related tools

  • Qwen3.5-122B-A10BAlibaba's flagship open-weight Qwen3.5 model — a 122B mixture-of-experts LLM with only ~10B active parameters, a hybrid DeltaNet/attention design, and a long context window that can run locally on high-memory Apple Silicon.
  • VibeThinker-3BA 3-billion-parameter open reasoning model that matches far larger models on math and code.
  • Gemma 4Google DeepMind's open-weight multimodal model family — reads images and audio, generates text, and runs fast and cheap.
  • ApertusA fully open, multilingual foundation model built in Switzerland for sovereign AI.

← All AI tools · Get started →