OpenOrange: should you fund it?
The two verdicts
Funding Advocate · Small cheque, paid out only on three conditions
Where I land now: the same call with tighter terms. I'd agree a small cheque but pay it out only once three things are met: that paying, staying customer from outside the network; profit per customer; the second co-founder named. Otherwise pass and look again in six months.
Funding Skeptic · Pass for now
On the public evidence available, I would pass for now and reopen on specified evidence, rather than recommend funding because the cheque is small. The strongest reason is not that OpenOrange is young or obscure. It is that the proposed advantages are largely available elsewhere, while customer preference, sustainable margins and operational quality remain unverified.
These are the Funding Advocate's and the Funding Skeptic's closing words from the debate.
What both sides agree on
- There is no traction evidence at all. No revenue, user counts, retention, margins, burn or runway were found in public, and none were supplied to this debate.
- Neither side says invest now. The Advocate says fund only once due diligence confirms traction, retention, execution and competitive position, and that the investment "should not go ahead until they are"; the Skeptic says pass at this stage.
- Both name the same missing proof: paying usage, retention, margins and why customers would pick OpenOrange over competitors.
Researcher's findings
$0OpenOrange public funding found
$153MRaised by two rivals in May 2026
$1.3BOpenRouter post-money valuation
Not publishedOpenOrange model markup
Facts
- No public funding. Searches of the web, Hacker News and GitHub found no announced round. That is either an early entry point or a sign no investor has committed yet; the record alone doesn't say which.
- No public traction. I found no published revenue, customer, user or retention figures.
- Pricing is usage-based. Free trial with no card; no subscription; prepaid balance; 100 Juice Units = $1; no fee on calls using the customer's own model keys; custom company terms by email.
- The margin is invisible. OpenOrange charges each model's price but does not publish its markup, so its gross margin can't be estimated from public information.
- No fee on own-key calls means customers who bring their own keys may generate little or no revenue per call.
Rival funding (USD millions)
| Company | Round | Amount | Announced | Post-money |
|---|---|---|---|---|
| OpenRouter | Series B | 113 | 2026-05-26 | 1,300 |
| Dust | Series B | 40 | 2026-05-19 | not disclosed |
| OpenOrange | None found | 0 | n/a | n/a |
| Rivals total | 153 |
Prices compared
| Product | Item | Price |
|---|---|---|
| OpenOrange | Free trial | $0, no card |
| OpenOrange | Subscription | $0 (prepaid balance only) |
| OpenOrange | Juice Unit | $0.01 (100 = $1) |
| OpenOrange | Fee on own-key calls | $0 |
| OpenOrange | Models it provides | Model price + unpublished markup |
| OpenOrange | Company terms | Custom, by email |
| Agent 37 | Managed OpenClaw, entry | $3.99 / month* |
| ClawHosters | Managed OpenClaw EU, entry | €19 / month* |
* Both rival prices come from Agent 37's own blog. Agent 37 is a competitor; treat these as unverified. Because OpenOrange bills by usage and rivals bill monthly, the rows do not compare one-for-one.
What would settle it
- Monthly revenue and active customers, with retention by cohort.
- Gross margin per Juice Unit and per deployment, including the unpublished model markup.
- Share of usage on customers' own keys (which earns no call fee).
- Burn rate and runway.
- Paying customers willing to take reference calls.
- A credible answer to why model and cloud providers can't copy it.
Sources
- TechCrunch: OpenRouter Series B, 2026-05-26
- The SaaS News: Dust raises $40M Series B
- openorange.ai/pricing
- openorange.ai/llms-full.txt (Juice Unit rate)
- Agent 37 blog: OpenClaw hosting providers 2026 (competitor source)
- OpenOrange funding: no result in web, Hacker News or GitHub searches.