Assets we generated, and how.
Work from our own generation pipelines, running on studio hardware. The method is on show alongside the output, because the method is the part worth hiring.
How we work with it
We do not hand a brief to a model and ship what comes back. The thinking is ours: what the thing needs to be, what good looks like, and which of forty outputs is the one worth keeping.
AI is an assistant here, not a substitute for the work. It is fast, tireless and confidently wrong, and it needs managing the way anything fast, tireless and confidently wrong needs managing. So we plan with it, direct it, correct it, and throw most of what it makes away.
What ships is what we chose, not what a model happened to produce.
Used this way it makes the work faster without making it thinner. The judgement is still the job; the assistant is what removes the hours around it.
Local language models
These run on studio hardware. Client material never leaves the building, and no third party is metering the work while we iterate — so a draft gets redone because it should be, not because a usage line allows it. Hosted models are in daily use too, where they are the better tool; what decides it is the job in front of us, not a policy.
- Input
- Unstructured text, screenshots, documents
- Models
- Gemma 4 12B QAT, gpt-oss-20b, Qwen3 8B, Gemma 4 E4B QAT
- Steps
- input, local model, schema check, structured output
Gemma 4 12B QAT
11.9B · LM Studio · OllamaA capable coder, and better than most of its size at planning a piece of work before any code exists. The one reached for most often.
Details
Google's multimodal model, trained with quantization built in rather than compressed afterwards — so it keeps close to full quality at roughly a quarter of the memory.
- Running well on hardware that should not really fit it
- Text and image in, 256k context
- Over 140 languages
gpt-oss-20b
21B total, 3.6B active · Ollama · Apache 2.0Agent-shaped work — calling tools, following a multi-step task, and showing its reasoning so a wrong turn can be traced. Quick for its size, and nothing leaves the machine.
Details
OpenAI's open-weight model. Mixture-of-experts, so it runs at the speed of a small model while reasoning like a bigger one.
- Tool use and function calling — the agent-shaped work
- Reasoning effort you can dial up or down per task
- Showing its full chain of thought, so a wrong answer is debuggable
Qwen3 8B
8.2B · OllamaQuick local work where a larger model would be overkill.
Details
A small general-purpose model with tool use and a thinking mode.
- Tool calls in a small package
- Showing its reasoning
- 40k tokens of context
Gemma 4 E4B QAT
Effective 4B · OllamaA genuinely good lightweight chatbot — small enough to stay quick, capable enough to be worth running rather than tolerated.
Details
The small one. Same quantization-aware training as its larger sibling, at a size that stays responsive on ordinary hardware.
- Holding a conversation without a wait between replies
- Running comfortably where a 12B model would not
- Text and image in, like the rest of the Gemma 4 family
hey - got your details off a mate. we run a small disability support service up in newcastle, about 18 staff. need a proper website, the current one is wix and honestly looks it. bigger thing though is staff training, we're doing inductions out of a folder and it's a mess. budget probably 15-20k but could stretch if the training side works out. no massive rush but ideally sorted before the new financial year. easiest to get me on the mobile in the arvo, mornings are chaos here
{
"intent": "new_project",
"organisation": {
"sector": "disability_support",
"location": "Newcastle, NSW",
"staff": 18
},
"services": ["company-website", "lms"],
"priority": "lms",
"replacing": "Wix",
"budget": { "min": 15000, "max": 20000, "flexible": true },
"deadline": { "date": "2026-06-30", "firmness": "preferred" },
"contact": { "channel": "phone", "window": "afternoon" },
"next_action": "call"
}- Retrieval
- nomic-embed-text v1.5 and qwen3-embedding, so a body of documents can be searched by meaning rather than by keyword. Also local.
- The line-up
- Not fixed. Document work ran on Nemotron until the quantization-aware Gemma 4 releases did the same job on less hardware. Running the models ourselves means changing one is a download rather than a contract.
- Hosted models
- Not everything runs here, and it was never meant to. Claude is in daily use across our own work — Haiku, Sonnet, Opus and Fable each suit a different size of task, and knowing which to reach for is most of the skill — with Gemini as the second option, for work that does not need Claude. Local and hosted are two tools rather than two camps: what settles it is whether the material can leave the building, and which model actually does the job better.
This page is the generation work. The interface side — the components we build with — is on The Workbench.