prism-ml
PrismML: Ternary Bonsai 2 27B
prism-ml/ternary-bonsai-2-27b
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
PrismML's Ternary Bonsai 2 27B is a 27-billion-parameter reasoning model derived from Qwen3.8-27B, distributed in GGUF format for llama.cpp and listed on OpenRouter with text-plus-image input and text output. Vendor documentation describes ternary weights near 1.7 bits per weight, shrinking the language model to roughly 6 GB while retaining a 262K-token context. Reasoning is enabled by default at high effort, and tool calling is supported. Reported quality and speed figures are vendor measurements, so independent validation is advised.[1][2]
Strengths
- Vendor documentation reports a roughly 6 GB language model with ternary weights at about 1.7 bits per weight, enabling 27B-class inference on a laptop or single GPU.[2]
- Long 262K-token context and optional image input via a separate vision pack that ships alongside the text weights.[1][2]
- Reasoning is on by default with adjustable effort, and the catalog lists tool calling, structured outputs, and logprobs among supported parameters.[1][2]
Limitations
- Evidence comes only from the OpenRouter catalog and the vendor's own documentation; quality and throughput figures are vendor-reported and were not independently verified, so validate on your own workload.[1][2]
- The densest packings require PrismML's llama.cpp fork; the stock-compatible file needs llama.cpp b11556 or newer, and older builds reportedly load it without warning while producing garbage output.[2]
- A low reasoning effort setting is not supported and behaves close to the default high effort; the knowledge cutoff is not stated in the supplied evidence.[1][2]
Where to use it
- May suit on-device or single-GPU assistants where a 27B-class reasoning model must fit in limited memory.[2]
- Could fit long-document or multimodal applications that need a large context window with occasional image input.[1][2]
- A candidate for agentic workflows that combine tool calling, structured outputs, and tunable reasoning effort, pending evaluation.[1][2]
Sources
- [1] OpenRouter catalog · PrismML: Ternary Bonsai 2 27BSources checked: Nov 13, 2026
- [2] Model documentation · PrismML: Ternary Bonsai 2 27BSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...
Specifications
- Model ID
- prism-ml/ternary-bonsai-2-27b
- Canonical ID
- prism-ml/ternary-bonsai-2-27b-20260918
- Creator
- prism-ml
- Added to OpenRouter
- 18.09.2026
- Context length
- 262,144
- Modality
- text+image->text
- Input modalities
- text, image
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.075 / out $0.500
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: optional · default xhigh · efforts: xhigh, medium, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Coding | No measurement | — |
| Agents | No measurement | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · default xhigh · efforts: xhigh, medium, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Document analysis | No measurement | — |
| Vision | No measurement | input: image |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Darkbloomint4 | 262,144 | $0.075 | $0.500 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.