Prism ML fits a 2B vision-language model into 0.52 GB
Prism ML’s ternary-weight Ternary-Bonsai-2-2B reduces the language model to about 0.52 GB, versus roughly 3.8 GB at FP16. It retains a 262K-token context window and image input, with the vision component loaded separately.
AlphaSignalNov 7, 2026
1 min read
Weights take values of −1, 0, or +1; additional scale metadata raises the actual storage cost. Image input requires an optional 0.36 GB Q8_0 mmproj file.
Derived from Qwen3.5-2B, the model is licensed under Apache 2.0. Most packings require PrismML’s llama.cpp fork; the source says only the Q2_0 file works with upstream llama.cpp b11556 and later.
Why it matters
The smaller file may help developers exploring local multimodal deployment on devices with tight storage and memory limits; performance on specific hardware still needs testing.
GUIDES
Something to try
As a hypothetical evaluation, compare supported packings on your target device and measure file size and inference-time memory use.
PUSULAM / UNDERSTAND
A good question changes everything.
Let’s unpack this article together. Ask what it means, why it matters, or how to use the idea.
A personal overviewFor your work and interests
What does this mean for you?
An explanation shaped around your work and interests.
Tell us a little about yourself to get an explanation that fits.
PUSULAM / COMMUNITY
The conversation
What did this change for you? Share a useful example or ask other readers a question.
Open-source Bark can generate speech, music, laughter, and sound effects from a single text prompt. The source says it supports 13 languages and more than 100 voice presets, but it may alter wording or add unexpected sounds. Bark generates audio clips of about 13–14 seconds and is released under the MIT license. The source reports about 12 GB of GPU memory for the full models, with smaller variants available for lower-memory use.
AutoTrust Labs released JEV-27B v1.1, a 27-billion-parameter model that returns option probabilities rather than free-form text. The company reports a score of 67.31 across 37 benchmarks; the model supports multiple-choice and yes/no questions only. The model selects among supplied options and returns a probability distribution. It stands out on tool selection and intent classification, but is weaker on open-ended reasoning and some other tests.
A review of 28 studies reports that agricultural TinyML deployments largely rely on ESP32, STM32, and ATMega microcontrollers. Some plant-disease detection systems report 90–97% accuracy and latency as low as 7.6 ms, but memory and energy metrics are often missing. The review says models typically run inference on-device in the field while training remains in the cloud. Quantization appears in over half of deployments, whereas structured pruning and neural architecture search are rare.
Start the conversation.