Australian Startup Accelerators: A Dated Program Finder

Choose a programme that addresses your next company constraint—not the most impressive investment headline. This guide is for Australian startup-curious founders comparing named sources. Wanting founder support does not mean you are looking for contractor work. If you have not decided whether a programme is useful at all, start with the accelerator decision explainer.

The directory covers generalist, AI, university, research, climate and inception-stage pathways. It is not exhaustive, a recommendation to invest, or evidence that every listed programme accepts your company. A provider's stated benefit is not a measured return on participation.

Compare the seven routes

As checked 29 October 2026. Follow each provider for current applications and full terms; a stated offer is conditional on selection and its documents.

See the worked choice or download the evidence worksheet. For local founder discussions, use the Melbourne founder-network guide and Sydney meetup guide.

Seven programme source records

Compare eligibility first, then the relevant support, costs and calendar. “Within published programme dates” means a schedule comparison—not independently confirmed activity or permission to join late. The former July snapshot is retained in the change history; it does not renew current details.

Some details remain unavailable. MAP's programme text was retrievable on 29 October, but its Key Dates banner supplied no readable dates; the September browser challenge is historical. Its former five-month duration is not retained as a current fact. Antler's eight-week overview and an older ten-week cohort news card should not be combined into a made-up next schedule. UNSW’s ended 2026 cohort is not relabelled “2027 applications”.

Compare cash, ownership and founder time separately

The Startmate page checked 29 October offers a SAFE or priced equity investment. The separate older terms guide explains a SAFE cap; confirm which instrument and terms apply to the actual offer. The simple ratio A$120,000 ÷ A$1.5 million = 8% is arithmetic for asking a question, not a complete ownership forecast. Other securities, financing and actual instrument terms matter.

UNSW's published uncapped discounted pre-money SAFE cannot produce a final ownership percentage from the cash and discount alone. Equity-free support is also not the same as unrestricted cash: product credits have separate eligibility and usage conditions. Programme support, investment selection and follow-on funding must not be collapsed into a guaranteed cheque.

• Cash: application/participation fees, travel, accommodation and expenses. Unknown amounts stay unknown.
• Founder time: application work, sessions, travel, preparation and displaced customer/product work. Hours are not automatically a cash expense.
• Rights and obligations: the full instrument, future rights, IP, attendance and any conditions. Ask suitably qualified advisers to review the actual offer.

This is general educational information, not legal, tax, financial or investment advice. Provider materials are not an independent assessment of what is appropriate for your company.

Worked choice: why this team defers

Fictional teaching case, not an MLAI client, applicant or interview. HarbourBrief is a Sydney AI quoting-workflow prototype. Both founders intend to retain full-time outside jobs. Their next question is whether a target buyer can provide suitable data; no buyer commitment, dataset permission or test outcome is invented.

At the 29 October source check, Startmate's full-time-founder requirement and three in-person weeks conflict with these assumptions. That is a reason to defer, even if its generalist support and investment headline look attractive. Submitting an EOI would not cure the mismatch. The team could investigate buyer/data access within its existing capacity and revisit programme selection if its commitment changes.

Illustrative planning arithmetic: two founders × four hours × twelve weeks = 96 founder-hours; adding eight application hours and 24 travel hours gives 128. These are hypothetical inputs, not a Startmate workload estimate, a measured result or a complete cost. Cash expenses and actual required hours are unknown.

Download the completed example and blank evidence worksheet (TXT). Open it in a text editor. It is a dated planning aid, not a saved application or a promise of current intake availability.

MLAI ACCELERATOR EVIDENCE RECORD
READER-ENTERED PLANNING RECORD — not independently verified.
Editorial checklist, not an admission/success predictor or legal/financial advice. No application is submitted.
Programme: Not entered
Cohort: Not entered
Source check: Not entered
Review trigger: Not entered
Evidence summary: 0 of 8 items supported by your entries. Clarify essential checks.

Support matches our actual constraint
Status: Unknown
Evidence / URL / uncertainty: Not recorded

Every eligibility requirement is met
Status: Unknown
Evidence / URL / uncertainty: Not recorded

Complete terms are understood and acceptable
Status: Unknown
Evidence / URL / uncertainty: Not recorded

Attendance and opportunity costs are acceptable
Status: Unknown
Evidence / URL / uncertainty: Not recorded

Relevant support and access are substantiated
Status: Unknown
Evidence / URL / uncertainty: Not recorded

Relevant alumni evidence has been assessed
Status: Unknown
Evidence / URL / uncertainty: Not recorded

A specific alternative has been compared
Status: Unknown
Evidence / URL / uncertainty: Not recorded

A decision rule and review trigger are recorded
Status: Unknown
Evidence / URL / uncertainty: Not recorded

Decision / reason: Not recorded
Use actual provider terms and appropriate independent advice. Keep private founder or alumni information out of shared copies.

Ask for experience without inventing proof of impact

Ask willing alumni at a comparable stage what support they actually used, what was missing, what work was displaced, and which terms they misunderstood. Record how they were recruited and any connection to the provider. Seek a range of experiences where available; do not assume a testimonial is typical or that a fixed number of interviews proves a return.

A portfolio logo shows participation, not causation. Ask what changed and what other explanations are plausible; an alumnus cannot establish with certainty what would have happened without the programme. Obtain permission before identifying people or sharing quotations/private terms. This guide contains no newly conducted alumni interviews.

A decision sequence, paced around the evidence

1. Name the company constraint and check whether a programme is the appropriate kind of support.
2. Eliminate known mismatches in eligibility, sector, commitment or timing; keep unknowns separate.
3. Request current terms, attendance and the specific support you need.
4. Seek relevant provider/alumni clarification when it could change the decision. Do not invent responses to fill a schedule.
5. Compare one realistic alternative using the same time and resources.
6. Record apply, clarify, defer or decline, with a reason and review trigger.

You might organise this over a week, but seven days is not an evidence-based deadline. Replies, advisers and founder availability determine the actual pace. Bring one non-confidential unresolved question to a relevant community conversation; event attendance does not guarantee alumni access, funding or admission.

Method, changelog, limitations and disclosure

Scope: seven selected provider-owned records connecting structured founder/research support to Australia. They are source leads, not endorsements, a complete census or an audited comparison of outcomes. Antler is retained as a distinct residency pathway, not relabelled a conventional accelerator.

Maintenance: MLAI editorial. Recheck by 30 October 2026, before the 1 November Startmate deadline, and whenever a provider reports a change. Source status uses the reader's calendar date on server render and updates in an open scripted tab. Invalid dates/clocks cannot certify a recent check. A review deadline does not perform a new check automatically.

29 October changes: checked all seven official programme pages. Startmate now lists 14 weeks, 1 February–7 May 2027 and a 1 November 2026 closing deadline; its former 8 November date is superseded. Added EnergyLab’s no-priced-round valuation condition; removed reliance on CSIRO’s no-longer-confirmed grant amounts; retained unknown MAP and Antler next-cohort dates. No private application or offer was accessed.

10 September changes: refreshed seven records, retained MAP's access limit, separated EOI/closing/start/phase facts, corrected the undated UNSW EOI and removed unverified duration/attendance assumptions. The July 29 baseline and September 9 Startmate/CSIRO intake corrections are historical snapshots; the previous CSIRO 31 August deadline is not reused as a new round.

Worksheet change: the previous 0–16 tally and “strong shortlist” threshold could outweigh missing eligibility. The replacement records evidence and unresolved requirements without an additive recommendation. Its worked case is fictional and its maths uses stated assumptions.

Review and assistance: AI tools assisted source discovery, drafting and implementation. No independent named programme/financial reviewer, private offer review, alumni interviews or causal outcome study is claimed. The illustration is not programme-attendance evidence. Send a source-backed correction if a provider detail has changed; use the provider for application questions.