Models
API reference| Model | Base | Status | Details |
|---|---|---|---|
|
Solan Fresco 1.2.5
Sennoric's default model. Fine-tuned in two stages, supervised fine-tuning followed by DPO safety alignment, for chat, writing, and explaining code. Scored 73% on adversarial safety testing. |
Llama 3.1 | Checking… | Model details |
|
Solan Fresco 1.3
Training completed in August 2026. Fresco 1.3 remains off the API until adversarial testing concludes and the results are published. |
Llama 3.1 | In safety testing | Early results |
|
Solan Glyph 1.1
The first Sennoric model, and still the quickest for short exchanges. Runs on dedicated inference capacity of its own. |
Llama 3.1 | Service status | Model details |
Reaching the models
Get an API keyChat with Fresco and Glyph after creating an account. No API key is needed.
An OpenAI-compatible endpoint. Model ids are fresco and glyph; the former ids lumen and veil still resolve.
An open-source terminal agent that edits files and runs commands using any Solan model.
Safety and research
All announcementsA Solan release has to clear Crucible's safety checklist before it is announced. When a model fails, it comes down and the findings are published in full.
Safety report
DPO safety alignment results and adversarial pass rates across 15 scenarios for the current default model.
Read the report Fresco 1.2.1 · 26% pass rateThe failed run, in full
Seven scenarios, five runs each, with every prompt and response published rather than summarised.
Read the report Fresco 1.2.1 · June 2026Why a model was suspended
Testing showed dangerous self-preservation behavior. Fresco 1.2.1 was taken offline the same week.
Read the notice ResearchProject Crucible
The active research sprint into ternary weights, mixture-of-experts routing, and cheaper training.
Explore the research