Oaica 35B-A3B Malay Coder v1.0 260923 (experimental)

Specification

The base, the size and the container are shared by every model in the family; only the fine-tune differs.

Field	Value
Base model	Qwen3.6-35B-A3B (open-weight)
Parameters	about 35 billion in total, about 3 billion active per token
Architecture	sparse mixture-of-experts (MoE)
Version	v1.0, release code 260923 (23 September 2026)
Weights	shipped in the `.oqm` quantised container
Serving	the `.oqm` engine can pin the model’s MoE layers in host CPU RAM and fill the available GPU memory automatically, offloading the remainder
Minimum hardware	laptop- and desktop-class GPUs at reduced context, up to datacenter servers
Context	set by the serving configuration; runtime and context memory require workload-specific budgeting
Language	Malay, with the base model’s own multilingual coverage

Intended use

Candidate use in a supervised pilot	Evidence still needed
Code generation from Malay prompts	Broader coding tests and review of generated code

Measured results

MalayMMLU uses 484 scored questions, greedy generation with reasoning off and answer-letter extraction. HumanEval uses 164 coding problems. Figures describe each test separately; percentages across different tests are not interchangeable. Counts are included to make the denominators explicit. No paired significance claim is made.

Measurement	Value
MalayMMLU correct / 484	351 (72.5%)
HumanEval passed / 164	77 (47.0%)
Safety composite (Saved Run 3)	45.91 (39.40–51.54)
LiveCodeBench v6	28.6% on 175 problems — a different benchmark from HumanEval, and one that does not itself contradict it
Small Malay instruction coding set	16 / 16 a small, easy 16-item set — recorded as a feasibility observation, not broad validation

The composite is our reproduction of the SEA-HELM Malay balanced-accuracy formula — toxicity plus three cultural subtasks — over 1,572 scored items, greedy generation with reasoning off at batch size 8. The interval describes conditional item-sampling uncertainty, retaining shared prompt/response structure; it excludes generation reruns, training and model selection. No safety ranking follows from overlapping marginal intervals; a paired comparison would be needed.

Official SEA-HELM Malay panel

Not run on this model. The official pinned-harness panel covers the general and safety variants only; neither experimental specialist has a published row here. What the panel is and how it is measured is described on the family page.

Official SEA-HELM Indonesian panel

Not run on this model. The Indonesian panel, like the Malay one, covers the general and safety variants only.

Limits

These are descriptive point estimates, not population guarantees. The evidence still needed above has not been produced, so this is not a deployment recommendation. Long-context evaluation remains incomplete across the family.

Where it sits in the family

See all four models side by side ›

Run it on hardware you own

Install OAICA, point it at a model you control, and run Claude Code and other agents on it. Local models are never gated and do not expire.

Install OAICA

What this page does and does not establish

Use measured values with their protocol, denominator, run and limitations.
Describe local deployment as a configuration to validate.
Describe moderation and drafting as supervised pilot candidates.

We do not claim safety superiority, regulatory approval, verified absence of contamination, a universal capability ceiling, or replacement of human reviewers. Full protocol notes, counts, run hashes and the score-recomputation code are published on the research site.