Full technical report PDF: https://huggingface.co/spaces/tri-fair-lab/publications/blob...

> In this report, we argue that frontier performance can be achieved by a wide range of institutions through Continual Learning on readily available open-weight models.

> As opposed to existing limited approaches such as small-scale fine-tuning, prompt engineering, or tool-augmentation with a frozen model, our Continual Learning approach takes advantage of the effectiveness of a modern mid- & post-training stack while introducing safeguards preserving both plasticity and stability at each training stage and seeking to make the minimal number of high-impact interventions on the parameters.

Benchmark overall score comparisons:

> Large: Thomson 1.0-Large - 78.5, Opus 4.8 - 79.5, Gemini 3.1 Pro - 78.0, GPT 5.4 - 76.5, Qwen3.5 397B - 73.0, Snowdon 1.0-Large - 73.6

> Small: Thomson 1.0-Small - 74.6, Gemma4 31B - 71.2, Snowdon 1.1-Small - 71.7, Qwen3.6 35B - 71.7

For the large model, Thomson is utilizing the fine tuning stack they describe in the article, running it on Snowdon 1.0-Large, which in turn is a fine tune of Qwen3.5 397B. For the small model, it's a fine tune of Snowdon 1.1-Small, which is a fine tune of Qwen3.6 35B.

As for the small version's run:

> The full pipeline consumed approximately 1.63 × 10²³ FLOP over 35,207 B200 GPU-hours, showing that these results are achievable with compute and personnel budgets substantially lower than commonly thought.

That would amount to around quarter to half a million dollars of spend on that run.

So it's a qwen fine-tune?

I mean that's a reasonable thing to do, but then the press release shouldn't be written the way it is written.

They're not as detached from the rest as the industry as the writing suggests.

__

> It is obtained by repurposing the open-weight Qwen3.6-35B-A3B model and substantially improving it on a wide range of performance domains.

nice wording on the HF page tho. "Repurposing". Lmao