The SoreQen S1 series

Three open-weight reasoning models at 0.8B, 2B and 4B parameters. S1 is where the line starts, not what it is — later generations will be more capable, and the direction does not change.

The three

Context and reply limits are what the API serves. A local build is limited by whatever you give it.
ModelParametersContextMax reply
SoreQen S1 Mini0.8B256k32k
SoreQen S12B256k32k
SoreQen S1 Mega4B256k32k

SoreQen S1 Mini

0.8Bsoreqen-s1-mini

Fastest. Good for quick questions and everyday chat.

Reasoning is a switch on this model rather than a dial. Its useful range is “don’t think”, “think a little” and “think harder”; asked for more, a 0.8B model spends its whole output allowance reasoning and returns no answer. That is measured, not assumed.

WeightsGGUF builds

SoreQen S1

2Bsoreqen-s1

The default. Balanced speed and depth.

WeightsGGUF builds

SoreQen S1 Mega

4Bsoreqen-s1-mega

Deepest reasoning. Best for hard problems and long documents.

WeightsGGUF builds

What every one of them can do

Capability is not a tier here. The chat product gates some of this by plan because a subscription is an allowance; the API does not, because a call is a purchase.

  • Vision — images in the message content
  • Tool calling, with results coming back as turns
  • Structured output against a strict JSON Schema
  • Streaming, with a usage frame at the end
  • Reasoning traces you can read
  • 256k of context on all three

Evaluation

One evaluation has been run and published: 24 Hinglish prompts, SoreQen S1 Mega against the Qwen3.5 4B base it was fine-tuned from, scored on measurable properties of the answer rather than on preference.

Out of 24 prompts each. Two of the five rows do not favour the fine-tune.
MeasureS1 MegaBase
Stayed in Roman scriptNeither model slipped into Devanagari. The product promises Roman-script Hinglish.2424
Genuinely code-mixedAt least three Hindi markers and two English content words in the same answer.2015
Carried sequence or causeWords like pehle, phir, kyunki — the difference between an answer and a greeting.2324
Drifted into EnglishAnswering a Hinglish question wholly in English. One case, against none.10
Produced the artefact asked forOf the three prompts that asked for one. Too few to read anything into.33
Hindi markers per answerMean across all 24. Higher means more actual code-mixing.22.521.1

What this does not tell you

Twenty-four prompts is a small sample, and one evaluation of one model is not a benchmark suite. It says the fine-tune measurably increased code-mixing — twenty answers against fifteen — and left informativeness roughly where it found it. It says nothing about maths, coding, long-context recall or tool use.

We have not run MMLU, GSM8K, HumanEval or any of the public leaderboards, so there are no scores for them on this page. A table of numbers nobody produced is the thing every model page has and the reason nobody believes any of them. When those runs happen, the results go here whichever way they come out.

Running them yourself

Every model is published in safetensors and in GGUF. Nothing needs to be signed into and nothing phones home. Two quantisations are built for each: Q4_K_M (the default. best size-to-quality trade; runs on a modest laptop) and Q8_0 (near-lossless. take it if you have the ram).

All three derive from Qwen3.5 (Alibaba Cloud, Apache 2.0). The full statement of what was changed, and under what terms each artefact is released, is on the open-source page rather than repeated here.