Crypto Keno Casinos Ranked by Real Paytable RTP
Eight casinos run their own in-house Keno. We opened every one of them, read the multiplier table at each risk level, and calculated the return from the exact 40/10 hit probabilities instead of trusting the number in the game description. The result is less glamorous than the marketing: five of the eight pay from one and the same table, a sixth matches it on three of its four levels, one pays more, one pays less, and the real differences live in payout limits, demo access and how each casino lets you check a draw.
Read this first if you are in the US or the UK
All eight casinos list the United States, the United Kingdom and the Netherlands as prohibited jurisdictions in their terms; seven of eight list Australia and France, six list Germany and Spain. If you are in one of those, the sign-up buttons below will not lead anywhere useful, and a VPN is a breach of the terms at all eight. Canada does not appear on any of the eight lists except Stake's, which names Ontario and Alberta. Each card ends with the list we read for that casino, because they differ: Stake names 45 jurisdictions, Rainbet eight.
The ranking: eight casinos, one shared table
Order is decided by calculated return first. Where casinos share an identical table, the tie is broken by a stated rakeback on the game, then by the payout limit — published where there is one, otherwise the largest single win shown on the casino's public board — then by whether you can try the game before registering.
| # | Casino | Paytable RTP, 10 picks (before limits and rebates) | Table | Payout limit · largest win shown | Demo without account | Verification |
|---|---|---|---|---|---|---|
| 1 | Duel | 99.94% – 99.99% | Own, above the market | None published · $1.02M shown | Yes | In-game seeds, rotate |
| 2 | Gamdom | 98.76% – 99.04% | Shared "Stake" table + instant rakeback to a stated 100% on the first $5,000 wagered daily | $1.25M published | No (members: $0 demo stated) | In-game seeds, verify |
| 3 | Stake | 98.76% – 99.04% | Shared "Stake" table | None published · $10M shown | No | In-game seeds, verify |
| 4 | Roobet | 98.76% – 99.04% | Shared "Stake" table | None published · three wins shown at $2.93M – $3.00M | No (members: $0 demo stated) | In-game panel |
| 5 | Rainbet | 98.76% – 99.04% (Extreme: 98.02%) | Shared "Stake" table + own 56/10 Extreme | $2M published · $2.66M shown | No | Verifier page with the code on it |
| 6 | Shuffle | 98.76% – 99.04% | Shared "Stake" table | None published · 999,999 USDT shown | No | In-game link |
| 7 | Razed | 98.30% – 99.01% | Shared "Stake" table on Low, Medium and High; Classic shows a lower 4-hit value | None published · $499,999.50 shown | No (game hidden) | In-game seeds, verify |
| 8 | BC.Game | 97.15% – 97.89% | "Stake" table × 0.99, rounded down | Hidden until login · $2.11M shown | No | In-game icon |
Limits, demo behaviour, fairness panels, win boards and terms were read on 19 October 2026. Multiplier rows were read the same day except Gamdom's (11 September 2026) and Stake's (our 6 July 2026 snapshot); the date of every row is listed in the method section. Roobet and Razed show their tables only to logged-in users, so those two were read from an account. Rakeback and "zero edge" offers change without notice and are listed as stated by the casino, not verified by us.
One table, five logos
Stake, Gamdom, Roobet, Rainbet and Shuffle pay exactly the same multipliers for 10 picks: 1.40× for three hits and 100× for ten on Classic, 3.50× for four hits and 1,000× for ten on High, and so on down the line. Run through the hypergeometric formula, those four tables return 99.04% (Classic), 98.76% (Low), 98.97% (Medium) and 99.01% (High) — at all five casinos. Razed shows the same rows on Low, Medium and High and differs on Classic in one cell, the four-hit multiplier (see the note under the chart). If a review tells you one of these "has better Keno odds" than another, it has not read the tables.
| Casino | Classic | Low | Medium | High | Top multiplier (10/10) |
|---|---|---|---|---|---|
| Duel | 99.96% | 99.99% | 99.99% | 99.94% | 1,200× (High) |
| Stake · Gamdom · Roobet · Rainbet · Shuffle | 99.04% | 98.76% | 98.97% | 99.01% | 1,000× (Low, Medium, High) |
| Razed | 98.30%* | 98.76% | 98.97% | 99.01% | 1,000× |
| Rainbet Extreme (56/10) | 98.02% | 10,000× | |||
| BC.Game | 97.74% | 97.15% | 97.67% | 97.89% | 989.89× (High) |
The chart is the whole ranking in one picture. Duel costs between 13 and 58 cents per $1,000 played; the five shared-table casinos sit between $9.63 and $12.40 regardless of the logo on the screen; BC.Game is roughly twice that. The range inside each bar is as large as the gap between neighbouring groups, so the risk level you choose matters about as much as which of the five identical casinos you choose — which is why the ranking separates them on limits and access rather than on odds. Figures are computed from the raw multipliers and rounded only at the end.
* Razed displays its Classic four-hit multiplier as 2.2× where the shared table has 2.25×. Its interface rounds to one decimal, so the true value may be 2.25× and the Classic return 99.04%; we show the figure as displayed and treat 98.30% as the displayed-table scenario, not as an established fact. Razed's position does not depend on it: with 2.25× it would tie the shared group and still rank last among them on the smallest win shown.
Duel: the only table that beats the market
Duel's multipliers sit a few percent above the shared table everywhere it matters — 3.51× against 3.50×, 64× against 63×, 1,200× against 1,000× for a perfect 10/10 on High. That lifts every level into the 99.94% – 99.99% band before any rebate. Duel also states a 0.1% base house edge on its Originals which it refunds instantly within a Zero Edge allowance; the allowance size is not published for Keno, so the ranking uses the paytable alone.
BC.Game: the shared table, minus one percent, rounded down
BC.Game's multipliers are the shared ones multiplied by roughly 0.99 and then cut to two decimals: 1.40× becomes 1.38×, 2.25× becomes 2.22×, 1,000× becomes 989.89×. The casino states a 2.00% house edge. Measured from the multipliers it shows the player, the edge is between 2.11% and 2.85% depending on the level, because every truncation goes against the player. It is the only casino here where the stated RTP and the displayed table disagree by more than a rounding error.
Rainbet Extreme: 10,000× on a bigger board
Rainbet's four standard levels are the shared table on the usual 40-number board. Its fifth level, Extreme, moves the game to 56 numbers with 10 drawn and headlines a 10,000× top prize. The board change matters more than the prize: hitting 10 of 10 on 56/10 is 1 in 35,607,051,480, against 1 in 847,660,528 on 40/10 — forty-two times rarer (the odds calculator gives the exact figure for any pick count and board). The Extreme table returns 98.02%, exactly what Rainbet's own rules state, so the mode is honest about its return; it is simply a different game from the one the other seven casinos run.
Where the shared table stops deciding
With the odds settled, the choice between Stake, Gamdom, Roobet, Rainbet, Shuffle and Razed comes down to four things we could actually read.
Payout limits: published and shown
Only Rainbet ($2,000,000) and Gamdom ($1,250,000) publish a maximum win for the game. For the others the only public evidence is the win board, and a board shows what has been paid, not what cannot be. Roobet's top three Keno wins all sit within 2.2% of $3,000,000; Shuffle's largest is 999,999 USDT; Razed's is $499,999.50 — a $1,930.50 bet at exactly 259×. Three results converging on a round figure look like a cap; one result on its own proves only that the win was shown. Stake's board shows $10,000,000 on a $20,000 bet, the largest Keno payout we found anywhere.
Trying before registering
Stake's game description promises a free demo; Shuffle's FAQ says you can play without a bet; Roobet writes "Betting $0.00 will enter demo mode" under its button. We clicked all of them logged out. Each opened a registration form. Duel is the only casino that dealt a demo round to a visitor, and its fairness panel admits that demo rounds are not provably fair. Gamdom and Roobet state a $0 demo mode for account holders; Razed hides the whole game until you log in.
Checking a draw yourself
All eight expose a server-seed hash, client seed and nonce. Rainbet's verifier page shows the result-generation code next to the seed fields, so you can read the HMAC logic on the spot rather than trust a "verified" badge; Stake documents equivalent code in its fairness help pages. Stake, Gamdom and Razed give a full seeds-and-verify panel inside the game; Duel, Roobet, Shuffle and BC.Game expose seeds and a rotate button, with verification through their help pages.
Identity checks and withdrawal friction
Shuffle requires a government ID before the first withdrawal. Rainbet gives 72 hours once it asks and caps deposits at €500 until you pass. Razed caps fiat withdrawals at $50,000 a month above your deposits. Stake, Roobet, Gamdom, Duel and BC.Game reserve the right to ask at any time and before a withdrawal, without a stated threshold. None of this shows up in the odds, and all of it shows up the first time you win properly.
Operator cards: who each casino is for
1 Duel
Highest paytable return · only true demo
Strengths
- Every risk level returns more than any other casino here — Low and Medium sit at 99.99% from the table alone.
- Stated 0.1% base edge refunded instantly inside a Zero Edge allowance.
- A visitor can play a full demo round before deciding; nobody else offers that.
- Hotkeys, autopick and an effective-edge readout on every bet card.
Weaknesses
- No published maximum bet or maximum payout — large wins rely on an unstated policy.
- Demo rounds use a separate, non-verifiable outcome path; real fairness checks need an account.
- The restricted list is longer than the usual core: it adds the UAE, Lithuania, Slovakia, Montenegro, Armenia, Jordan and Kuwait.
Best for: players who pick a Keno by house edge first and want to see the game before registering.
Not for: anyone who needs a written payout cap before betting large, or who is in the restricted list below.
Not available in: United States, United Kingdom, Australia, Germany, France, Netherlands (incl. Curaçao), Spain, UAE, Lithuania, Slovakia, Montenegro, Armenia, Jordan, Kuwait and sanctioned countries.
2 Gamdom
Shared table, rakeback to a stated 100%, published limits
Strengths
- The instant rakeback is the only mechanism in the shared-table group that changes the effective return, and it covers the stake sizes most people play at.
- Clear, published bet range and payout cap on the game page — rare in this group.
- Risk slider, autobet with stop-on-profit and stop-on-loss, instant mode.
- The 37-entry restricted list does not name Canada, Italy, Poland, Czechia, Brazil, Mexico or India.
Weaknesses
- The paytable itself is the shared one; the "100% RTP" headline is a rebate, not better odds, and it stops after $5,000 of daily wagering.
- The game page contradicts itself on the maximum win ($1,250,000 in the spec table, "three million" in the FAQ); we use the lower figure.
- No demo for visitors; the bet-zero demo is offered to members only.
Best for: daily players at small and mid stakes who will actually use the $5,000 rakeback allowance.
Not for: 10/10 jackpot hunters — the $1.25M cap is below Rainbet's published $2M and below the wins shown at Stake and Roobet — and anyone in the 37 blocked jurisdictions.
Not available in: United States, United Kingdom, Australia, Germany, France, Netherlands, Spain, Austria, Sweden, Portugal, Denmark, Belgium, Greece, Malta, Hungary, Slovakia, Serbia, Slovenia, Lithuania, Luxembourg, Guernsey, Jersey, Aruba, Bonaire, Curaçao, Iran, North Korea and others (37 in total).
3 Stake
The reference table · biggest payouts on record
Strengths
- The largest Keno payouts we found anywhere: $10,000,000, $8,354,095 and $8,000,000 on the public board.
- Around a thousand people in the game at the moment we checked and 53,520 saved games — the deepest player pool of the eight.
- Full Fairness panel with an explained commitment scheme and verify tab.
Weaknesses
- No demo for visitors even though the game text promises one.
- The longest prohibited list here: 45 jurisdictions including Brazil, Mexico, Italy, Poland, Czechia, Argentina, Colombia, Peru and the Canadian provinces of Ontario and Alberta.
- No stated rakeback on the game; the VIP programme is where returns live, and it is not quantifiable from outside.
Best for: high-stakes play and 10-spot swings, on the evidence of the board: $20,000 bets and seven- and eight-figure payouts are shown there, and no other casino here shows anything close.
Not for: players in the Americas outside the open provinces of Canada, or anyone who wants to test the game first.
Not available in: United States, United Kingdom, Australia, Canada (Ontario, Alberta), Germany, France, Netherlands, Spain, Italy, Brazil, Mexico, Argentina, Colombia, Peru, Poland, Czech Republic, Sweden, Denmark, Portugal, Austria, Belgium, Greece, Croatia, Lithuania, Slovakia, Serbia, Malta, Cyprus, Israel, South Africa and others (45 in total).
4 Roobet
Shared table · $3M wins on the board · short restricted list
Strengths
- Second-largest wins in the group: the public board shows three separate players at $2.93M – $3.00M on single Keno bets.
- Takes very large stakes — the record wins came from bets above $36,000.
- A 25-entry restricted list; in Europe it names only the UK, the Netherlands, Cyprus, Malta and Gibraltar.
- Live RTP page and a Fun Mode for account holders.
Weaknesses
- "Betting $0.00 will enter demo mode" is shown to visitors, but the button opens registration until you have an account.
- Nothing about bet limits or payout caps is written down; the $3M figure is an inference from three similar wins on the board, not a rule.
- No rakeback or rebate stated for Keno.
Best for: mid-to-high stakes players outside its 25-entry list who want a Stake-equivalent table at a casino whose board shows multi-million Keno wins.
Not for: players who need limits in writing before they deposit.
Not available in: United States, United Kingdom, Australia, Netherlands, Cyprus, Malta, Gibraltar, Aruba, Bonaire, Curaçao, Saba, Sint Maarten, Statia, Cuba, Haiti, Nicaragua, Iran, Iraq, Syria, Yemen, Myanmar, North Korea, South Sudan, Zimbabwe.
5 Rainbet
Verifier with the code on the page · published limits · Extreme 56/10 mode
Strengths
- The verifier page shows the outcome code beside the seed fields — the most direct check of the eight; you can read how the HMAC turns seeds into drawn numbers without leaving the form.
- Every limit is in the rules: minimum, maximum, payout cap, and a per-level RTP statement that our calculation reproduces (Extreme to within 0.02 points).
- Shortest restricted list of the eight (8 jurisdictions).
- Extreme mode exists for players who want a 10,000× label and understand what it costs in odds.
Weaknesses
- $4,000 maximum bet rules out high-roller sessions.
- Extreme's 56-number board makes 10/10 forty-two times rarer than on 40/10; the headline multiplier is not comparable with the others.
- No demo in any form; the "i" panel and rules are the only preview.
- Rules and win board disagree on limits ($4,000 / $2,000,000 written against $5,318 / $2,659,220 shown), and nothing we read says which is current.
Best for: players who verify rounds themselves and bet in the $0.10 – $4,000 range.
Not for: high rollers, and for anyone reading 10,000× as a realistic target.
Not available in: United States, United Kingdom, Australia, Germany, France, Netherlands (incl. Aruba, Curaçao, Sint Maarten), Spain, Austria.
6 Shuffle
Shared table · largest win shown 999,999 USDT · ID before first withdrawal
Strengths
- Identical odds to Stake at every level; the game page states a 1.00% house edge, which sits inside the calculated 98.76% – 99.04% band.
- Fast rounds, a clean risk selector and the usual seed rotation.
- Restricted list is short (14) and does not name Canada, Italy, Poland, Brazil or most of Asia.
Weaknesses
- No win above 999,999 USDT appears on the board; whether that is a cap or simply the record so far is not written anywhere.
- Identity check is mandatory before you withdraw anything, not just above a threshold.
- The game description is lottery boilerplate ("20 winning numbers are drawn") and promises free play that does not exist for visitors — read the game screen, not the text.
Best for: players already banking on Shuffle who want Stake's Keno odds without Stake's restricted list.
Not for: anyone chasing a seven-figure hit, or anyone who wants to withdraw once without sending a passport.
Not available in: United States, United Kingdom, Australia, Germany, France, Netherlands, Spain, Portugal, Sweden, Singapore, Cyprus, Curaçao, British Virgin Islands, Cayman Islands.
7 Razed
Shared table · largest win shown $500K · nothing visible before login
Strengths
- Low, Medium and High are the shared table, so the odds are market-standard once you are inside.
- Seeds-and-verify panel with a per-pair bet counter and seed rotation.
- Accepts BTC, ETH, LTC and USDT with a USD display.
Weaknesses
- You cannot see the board or the paytable without an account — the worst pre-registration transparency of the eight.
- Smallest record of the eight: the largest win shown is $499,999.50, and no limit is published.
- Classic shows 2.2× for four hits where the shared table pays 2.25×; if that display is literal, Classic returns 98.30%. Its rank does not depend on it.
- Fiat withdrawals capped at $50,000 per month above deposits.
Best for: existing Razed account holders playing Low, Medium or High at modest stakes.
Not for: anyone comparing casinos before signing up, or anyone whose target win has seven figures.
Not available in: United States, United Kingdom, France and overseas territories, Netherlands, Austria, Israel, Singapore, Vietnam, Kazakhstan, Myanmar, Afghanistan, Iran, North Korea, Comoros, Aruba, Bonaire, Curaçao, Saba, Sint Eustatius and sanctioned countries.
8 BC.Game
Lowest return of the eight · micro-stakes
Strengths
- Bets from $0.0001 — the smallest minimum in the group, useful for testing a pick count with real draws.
- A $2,114,893 win on a $26,214 bet is shown on the public board.
- Autobet with percentage step-ups and stop limits; pick-count slider rather than clicking tiles one by one.
Weaknesses
- Every multiplier is the shared one minus about 1%, then rounded down: the measured edge is 2.11% – 2.85% against a stated 2.00%.
- Over 1,000 rounds at $10, the difference to the shared table is roughly $110 – $160 of expected loss; against Duel it is about $205 – $285.
- Maximum win and bet range read "--" and "$0 to $0" until you log in.
- Adds mainland China, Hong Kong and Macau to the usual restricted list.
Best for: micro-stake sessions and players who are already on BC.Game for other reasons.
Not for: anyone choosing a Keno by return — six casinos above pay more for the same picks.
Not available in: United States, United Kingdom, Australia, Austria, Germany, France, Netherlands, Spain, China (mainland), Hong Kong, Macau, Comoros, FATF-blacklisted countries.
How the numbers were produced
Each casino's Keno was opened on 19 October 2026, ten tiles were selected, and the multiplier row was recorded at every risk level the game offers. Three exceptions are dated separately in the table below: Stake's four rows come from our 6 July 2026 snapshot (the game, its description, fairness panel and board were re-read on 19 October); Gamdom's four rows were recorded on 11 September 2026 for our Gamdom calculator, with the board re-checked on 19 October; BC.Game's 9- and 10-hit cells come from the July snapshot because the October screen cut off there. Roobet and Razed display the table only to members, so those were read from a logged-in account; everything else was read as a visitor. The return of each table is the sum over hit counts of probability × multiplier:
Rainbet Extreme: P(k) = C(10,k) · C(46,10−k) / C(56,10)
The 33 multiplier rows behind this page (10 picks; hits 0 – 10, multiplier of stake)
| Casino | Level | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | RTP | Read |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Duel | Classic | 0 | 0 | 0 | 1.4 | 2.27 | 4.6 | 8.2 | 18 | 55 | 100 | 200 | 99.96% | 19 Oct 2026 |
| Duel | Low | 0 | 0 | 1.11 | 1.22 | 1.31 | 1.81 | 3.6 | 14 | 60 | 300 | 1,200 | 99.99% | 19 Oct 2026 |
| Duel | Medium | 0 | 0 | 0 | 1.61 | 2.01 | 4.1 | 7.1 | 27 | 110 | 600 | 1,200 | 99.99% | 19 Oct 2026 |
| Duel | High | 0 | 0 | 0 | 0 | 3.51 | 8.1 | 13.1 | 64 | 600 | 1,000 | 1,200 | 99.94% | 19 Oct 2026 |
| Gamdom | Classic | 0 | 0 | 0 | 1.4 | 2.25 | 4.5 | 8 | 17 | 50 | 80 | 100 | 99.04% | 11 Sep 2026 |
| Gamdom | Low | 0 | 0 | 1.1 | 1.2 | 1.3 | 1.8 | 3.5 | 13 | 50 | 250 | 1,000 | 98.76% | 11 Sep 2026 |
| Gamdom | Medium | 0 | 0 | 0 | 1.6 | 2 | 4 | 7 | 26 | 100 | 500 | 1,000 | 98.97% | 11 Sep 2026 |
| Gamdom | High | 0 | 0 | 0 | 0 | 3.5 | 8 | 13 | 63 | 500 | 800 | 1,000 | 99.01% | 11 Sep 2026 |
| Stake | Classic | 0 | 0 | 0 | 1.4 | 2.25 | 4.5 | 8 | 17 | 50 | 80 | 100 | 99.04% | 6 Jul 2026 snapshot |
| Stake | Low | 0 | 0 | 1.1 | 1.2 | 1.3 | 1.8 | 3.5 | 13 | 50 | 250 | 1,000 | 98.76% | 6 Jul 2026 snapshot |
| Stake | Medium | 0 | 0 | 0 | 1.6 | 2 | 4 | 7 | 26 | 100 | 500 | 1,000 | 98.97% | 6 Jul 2026 snapshot |
| Stake | High | 0 | 0 | 0 | 0 | 3.5 | 8 | 13 | 63 | 500 | 800 | 1,000 | 99.01% | 6 Jul 2026 snapshot |
| Roobet | Classic | 0 | 0 | 0 | 1.4 | 2.25 | 4.5 | 8 | 17 | 50 | 80 | 100 | 99.04% | 19 Oct 2026, logged in |
| Roobet | Low | 0 | 0 | 1.1 | 1.2 | 1.3 | 1.8 | 3.5 | 13 | 50 | 250 | 1,000 | 98.76% | 19 Oct 2026, logged in |
| Roobet | Medium | 0 | 0 | 0 | 1.6 | 2 | 4 | 7 | 26 | 100 | 500 | 1,000 | 98.97% | 19 Oct 2026, logged in |
| Roobet | High | 0 | 0 | 0 | 0 | 3.5 | 8 | 13 | 63 | 500 | 800 | 1,000 | 99.01% | 19 Oct 2026, logged in |
| Rainbet | Classic | 0 | 0 | 0 | 1.4 | 2.25 | 4.5 | 8 | 17 | 50 | 80 | 100 | 99.04% | 19 Oct 2026 |
| Rainbet | Low | 0 | 0 | 1.1 | 1.2 | 1.3 | 1.8 | 3.5 | 13 | 50 | 250 | 1,000 | 98.76% | 19 Oct 2026 |
| Rainbet | Medium | 0 | 0 | 0 | 1.6 | 2 | 4 | 7 | 26 | 100 | 500 | 1,000 | 98.97% | 19 Oct 2026 |
| Rainbet | High | 0 | 0 | 0 | 0 | 3.5 | 8 | 13 | 63 | 500 | 800 | 1,000 | 99.01% | 19 Oct 2026 |
| Rainbet | Extreme (56/10) | 0 | 0 | 0 | 0 | 3.28 | 44 | 300 | 1,500 | 5,000 | 8,000 | 10,000 | 98.02% | 19 Oct 2026 |
| Shuffle | Classic | 0 | 0 | 0 | 1.4 | 2.25 | 4.5 | 8 | 17 | 50 | 80 | 100 | 99.04% | 19 Oct 2026 |
| Shuffle | Low | 0 | 0 | 1.1 | 1.2 | 1.3 | 1.8 | 3.5 | 13 | 50 | 250 | 1,000 | 98.76% | 19 Oct 2026 |
| Shuffle | Medium | 0 | 0 | 0 | 1.6 | 2 | 4 | 7 | 26 | 100 | 500 | 1,000 | 98.97% | 19 Oct 2026 |
| Shuffle | High | 0 | 0 | 0 | 0 | 3.5 | 8 | 13 | 63 | 500 | 800 | 1,000 | 99.01% | 19 Oct 2026 |
| Razed | Classic | 0 | 0 | 0 | 1.4 | 2.2 | 4.5 | 8 | 17 | 50 | 80 | 100 | 98.30% | 19 Oct 2026, logged in |
| Razed | Low | 0 | 0 | 1.1 | 1.2 | 1.3 | 1.8 | 3.5 | 13 | 50 | 250 | 1,000 | 98.76% | 19 Oct 2026, logged in |
| Razed | Medium | 0 | 0 | 0 | 1.6 | 2 | 4 | 7 | 26 | 100 | 500 | 1,000 | 98.97% | 19 Oct 2026, logged in |
| Razed | High | 0 | 0 | 0 | 0 | 3.5 | 8 | 13 | 63 | 500 | 800 | 1,000 | 99.01% | 19 Oct 2026, logged in |
| BC.Game | Classic | 0 | 0 | 0 | 1.38 | 2.22 | 4.45 | 7.91 | 16.82 | 49.49 | 79.19 | 98.98 | 97.74% | 19 Oct 2026 (9–10 hits: 6 Jul 2026) |
| BC.Game | Low | 0 | 0 | 1.08 | 1.18 | 1.28 | 1.78 | 3.46 | 12.86 | 49.49 | 247.47 | 989.89 | 97.15% | 19 Oct 2026 (9–10 hits: 6 Jul 2026) |
| BC.Game | Medium | 0 | 0 | 0 | 1.58 | 1.97 | 3.95 | 6.92 | 25.73 | 98.98 | 494.94 | 989.89 | 97.67% | 19 Oct 2026 (9–10 hits: 6 Jul 2026) |
| BC.Game | High | 0 | 0 | 0 | 0 | 3.46 | 7.91 | 12.86 | 62.36 | 494.94 | 791.91 | 989.89 | 97.89% | 19 Oct 2026 (9–10 hits: 6 Jul 2026) |
Five casinos produced the same four tables to the second decimal, which is why they share one row above; Razed matched three of them and showed a different four-hit value on Classic. Seven-pick tables for Stake, Duel and BC.Game from a July 2026 snapshot were used only to confirm that the shared-table pattern holds at other pick counts; they do not enter the ranking.
- Ranking order: calculated 10-pick RTP across levels; identical tables are separated by a stated rakeback on the game, then by the payout limit (published, otherwise the largest win shown), then by demo access, then by verifier transparency.
- Limits marked "stated" come from the casino's rules or game page. "Shown" is the largest single Keno win on the public win board and is a lower bound on what the casino pays, not a limit; where several wins converge on one round figure we say so and call it an inference.
- Demo was tested by clicking the bet button while logged out, after selecting numbers.
- Not measured: withdrawal speed, support quality, bonuses and wagering contribution for Keno. None of the eight terms documents states a Keno contribution rate; those live on promotion pages that change weekly.
Multiplier tables, limits and restricted lists change without notice. The dates in this section are the dates each figure was read; a figure that matters to your bet should be confirmed on the game screen before you place it.
Questions people ask before choosing
Which crypto casino has the best Keno odds?
By the paytable, Duel: 99.94% – 99.99% at 10 picks across its four levels. Stake, Gamdom, Roobet, Rainbet and Shuffle are tied at 98.76% – 99.04% because they use one and the same multiplier table; Razed matches it on three levels and shows 98.30% on Classic as displayed; Gamdom's stated rakeback lifts its effective return within a daily allowance. BC.Game is last at 97.15% – 97.89%.
Why do five different casinos have identical Keno paytables?
In-house "Originals" Keno is a small genre and one multiplier set dominates it. Stake, Gamdom, Roobet, Rainbet and Shuffle pay the same multipliers at every hit count and level for 10 picks; Razed matches on Low, Medium and High; BC.Game uses the same set scaled down by about 1%. Why they coincide is not something a paytable can tell you — only that they do. Duel's table is different at every level.
Is "100% RTP" at Gamdom or Duel real?
It is a rebate, not a table. Gamdom's paytable returns 98.76% – 99.04%; the casino refunds the house share instantly on the first $5,000 you wager each day and states 99% after that. Duel's tables return 99.94% – 99.99% on their own and the casino refunds a stated 0.1% base edge within a Zero Edge allowance. In both cases the draw odds and multipliers are unchanged; the money comes back as a credit.
Can I play crypto Keno for free before depositing?
Without an account, only on Duel, and Duel says plainly that demo rounds are not provably fair. Gamdom and Roobet offer a $0 demo round to members, according to their own game screens. Stake and Shuffle promise free play in their text but open a registration form when you click; Razed does not show the game at all to visitors.
Rainbet Extreme pays 10,000×. Is it the best jackpot mode?
It is the biggest label, not the best chance. Extreme runs on 56 numbers instead of 40, so a perfect 10/10 is 1 in 35.6 billion rather than 1 in 848 million. The table returns 98.02%, slightly below the standard levels. A High-level 10/10 at 1,000× on a 40-number board is a far more reachable target.
Can players in the United States use any of these?
No. All eight list the United States in their prohibited jurisdictions, as they do the United Kingdom and the Netherlands; Australia and France appear on seven lists, Germany and Spain on six. The lists differ beyond that — Stake names 45 jurisdictions, Rainbet 8 — and each card above ends with the list we read for that casino.
Which one suits small stakes?
BC.Game takes bets from $0.0001, Rainbet and Gamdom from $0.10. At those sizes the paytable difference is pennies per round, so the better questions are whether you can verify draws (Rainbet) and whether a rakeback applies (Gamdom).
Does picking fewer numbers change the return?
Yes — each pick count has its own multiplier row and its own RTP, and the casinos differ in how far the rows deviate. This page ranks on 10 picks; the Keno strategy guide covers how pick count and risk level shift the return within one casino.
Responsible play: a higher paytable return lowers the expected cost of each round; it does not make any round more likely to win. Keno draws are independent, and no casino, risk level or pick count on this page changes that. Operator buttons are sponsored affiliate links.