Ruling — Rummikub

← Rummikub

You asked

How do I play?

Answered counted as answered

Asked with: Rummikubbg-811-en-1e

1. Place tiles face down on the table and mix them thoroughly.[1]

2. Each player picks a tile; the player with the highest number goes first.[1]

3. Return the tiles to the table and mix them.[1]

4. Each player takes 14 tiles and places them on his/her rack.[1]

5. To make an initial meld, each player must place tiles on the table in one or more sets that total at least 30 points using tiles from their rack; players may not use tiles already played on the table for this initial meld.[2]

6. On turns after the initial meld, a player can build onto other sets on the table with tiles from their rack.[2]

7. Players can manipulate sets on the table by rearranging or adding to them, such as splitting a run, adding tiles to make new sets, or removing tiles to form new sets, provided that at the end of the round only legitimate sets remain and no loose tiles are left.[3]

8. A joker can be used as any tile in a set.[4] A player who has already made their initial meld can retrieve a joker from a set on the table by replacing it with any tiles that keep the set legitimate, but they must play the joker on that same turn to make a new set and must also use at least one tile from their rack.[4]

9. If a player cannot play any tiles from their rack, or chooses not to, they must draw a tile from the pool, and their turn ends.[2] Players cannot play a tile they just drew until their next turn.[2]

10. There is a time limit of 1 minute per player, per turn.[5] If a player exceeds this limit, they must draw a tile from the pool and their turn ends.[5] If they cannot complete a move within the limit, they must replace the tiles to their previous positions, take back the tiles they played, and draw 3 tiles from the pool as a penalty.[5]

11. Play passes to the left (clockwise).[2]

12. The game ends when one player empties their rack and calls "Rummikub!".[2] If there are no more tiles in the pool and no player has emptied their rack, play continues until no more plays can be made.[2] In the rare case that all tiles in the pool are used before anyone calls "Rummikub!" and no more plays can be made, the player with the lowest value of tiles on their rack wins the round.[5]

[1] RULEBOOK Rummikub — Rulebook · p.1 · Assembly instructions: · Contents for Rummikub XP/Mini XP · Object of The Game: · Sets: · Set-up:

[2] RULEBOOK Rummikub — Rulebook · p.1 · Playing The Game:

[3] RULEBOOK Rummikub — Rulebook · p.2 · Manipulation: · Add one or more tiles from rack to make new set: · Splitting a run:

[4] RULEBOOK Rummikub — Rulebook · p.3 · The Joker: · 4 ways to clear the joker are:

[5] RULEBOOK Rummikub — Rulebook · p.4 · 4 ways to clear the joker are: · Time Limit: · Incomplete Runs: · Winning: · Scoring:

Answer in hand — your move.

🔧 debug — where this ruling came from

outcome: ANSWERED — Answered
corpus profile: live,vps
answerer pin (deploy-frozen, attribution only): gemma4:26b think:false 5571076f3d70
timings: search 82 ms · ask 5735 ms · total 5819 ms
single-flight (PR-5): queued 0 ms · generation 5819 ms
retrieval refs (raw) — what the PASSAGE SEARCH returned, not what the answerer read: [{"page": 1, "score": 0.04918, "snippet": "Each tile is worth its face value (the number shown on the tile). In order to make an initial meld, each player must place tiles on the table in one or more sets that total at least 30 points. These points must come from the tiles on each player's rack; for their initial meld, players may not use ti", "headingPath": "Playing The Game:", "sharpsignalDocId": 152}, {"page": 4, "score": 0.048131, "snippet": "| Sample score table: | Player A | Player B | Player C 1 | Player D | | Game 1 | 24 + | 15 - | 16 - | 43 - | | Game 2 | 26 - | 11 - | 22 + | 45 - | | Game 3 | 32 - | 13 - | 22 - | 47 + | | Total Round's | 14 - | 29 - | 4 + | 39 + | The beginning of a game of Rummikub® may seem slow, but as the table", "headingPath": "Strategy:", "sharpsignalDocId": 152}, {"page": 3, "score": 0.047371, "snippet": "There are two jokers in the game. Each joker can be used as any tile in a set, and its number and color are that of the tile needed to complete the set. On future turns, a joker can be retrieved from a set on the table by a player who can replace it during his/her turn with any tiles that can keep t", "headingPath": "The Joker: · 4 ways to clear the joker are:", "sharpsignalDocId": 152}, {"page": 4, "score": 0.046635, "snippet": "The player splits the run. He moves the black 1 to the group of ones, he moves the black 2 to the group of twos and frees the joker. Tiles on table There is a time limit of 1 minute per player, per turn. Players who go over the time limit must draw a tile from the pool, ending their turns. Players w", "headingPath": "4 ways to clear the joker are: · Time Limit: · Incomplete Runs: · Winning: · Scoring:", "sharpsignalDocId": 152}, {"page": 1, "score": 0.046432, "snippet": "1 2 3 Contents: 106 tiles (8 sets of tiles 1-13 in four colours, and 2 joker tiles), 4 racks + rack holders . for 5 or 6 players: use all of the 160 tiles (12 sets of tiles 1-13 in four colours, and 4 joker tiles), 6 racks + rack holders. To be the first player to play all the tiles from your rack b", "headingPath": "Assembly instructions: · Contents for Rummikub XP/Mini XP · Object of The Game: · Sets: · Set-up:", "sharpsignalDocId": 152}, {"page": 2, "score": 0.045921, "snippet": "Manipulation is the most exciting part of playing \"Rummikub ® \". Players try to table the greatest amount of tiles by rearranging or adding to sets which are already on the table. Sets can be manipulated in many ways (examples follow) as long as at the end of each round only legitimate sets remain a", "headingPath": "Manipulation: · Add one or more tiles from rack to make new set: · Splitting a run:", "sharpsignalDocId": 152}]
citations (raw) — what the answer cited, chosen from a retrieval this page cannot show: [{"page": 1, "quote": null, "headingPath": "Assembly instructions: · Contents for Rummikub XP/Mini XP · Object of The Game: · Sets: · Set-up:", "sharpsignalDocId": 152}, {"page": 1, "quote": null, "headingPath": "Playing The Game:", "sharpsignalDocId": 152}, {"page": 2, "quote": null, "headingPath": "Manipulation: · Add one or more tiles from rack to make new set: · Splitting a run:", "sharpsignalDocId": 152}, {"page": 3, "quote": null, "headingPath": "The Joker: · 4 ways to clear the joker are:", "sharpsignalDocId": 152}, {"page": 4, "quote": null, "headingPath": "4 ways to clear the joker are: · Time Limit: · Incomplete Runs: · Winning: · Scoring:", "sharpsignalDocId": 152}]
boost provenance — errata + base rulebook (E2/E3): no registered errata card matched this query; base rulebook doc 152 boosted; rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base rulebook doc 152 boosted
answered-before recall: this ruling was answered, so past rulings were not offered
similarity threshold in force: 0.55 (rulesage.answered-before.similarity-threshold; the exact-question tier is not gated by it)
answer fence: passed — letter density 0.92465 of 1765 non-space chars (floor 0.45); 0 glyph-index tokens (max 2)
debug row recorded: Aug 25, 2026 16:58 UTC

retrieval — what the passage search returned

what this block IS: the PASSAGE SEARCH's own results — the same search that fills this page's closest-passages list — and NOT the passages the answerer read. Answering runs its OWN retrieval: a far wider candidate pool, re-ranked down to a final handful, plus floors this search never applies — the base rulebook, surviving errata corrections, and one slot per book on a multi-book read. Each of those armed only if THIS ask sent what it needs: the boost block names the ids it forwarded, and the scope line names the books it read. A document the answer CITES that no row here names is therefore EXPECTED, not a defect: two searches, two candidate pools, two orders. The answerer's own final rows are shown in their own block below, beside these and never mixed into them.

score = the fused retrieval score (reciprocal-rank fusion over the search arms) — NOT a cosine similarity. This ruling ran the THREE-ARM fuse (full-text + vector + the doc-priority boost arm, which carried the base rulebook and/or the errata corrections this query matched), whose arithmetic ceiling is 0.04918 — all three arms ranking the same chunk #1 (ceiling = arms/(k+1) at k=60). The measured healthy band 0.016–0.033 was measured on TWO-arm asks and does not bound this one: its high end is the two-arm ceiling, and a third arm has room above it. A chunk only one arm ranked highly lands near 0.016 (that one arm's #1 vote), so 0.03 here is an excellent hit, not a 3% match.

per-arm ranks (the rank each search arm gave a chunk): NOT DISCLOSED by the gateway. Fusion happens gateway-side and only the FUSED score crosses the wire, so rulesage cannot show them without a gateway change. The rank column below is the FUSED rank (position in the returned list), which is real.

token count per chunk: NOT DISCLOSED — chunk token counts are not reported on the ask path. The character counts below are of the bounded snippet/quote rulesage actually received — a real measurement of what we hold, never an estimate of the chunk.

fused rank	document	heading	page	fused RRF score	snippet chars
#1	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Playing The Game:	p.1	0.04918 (at the 3-arm ceiling (0.04918) — every arm ranked it #1)	300
#2	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Strategy:	p.4	0.048131 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#3	RULEBOOK Rummikub — Rulebook (corpus doc 152)	The Joker: · 4 ways to clear the joker are:	p.3	0.047371 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#4	RULEBOOK Rummikub — Rulebook (corpus doc 152)	4 ways to clear the joker are: · Time Limit: · Incomplete Runs: · Winning: · Scoring:	p.4	0.046635 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#5	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Assembly instructions: · Contents for Rummikub XP/Mini XP · Object of The Game: · Sets: · Set-up:	p.1	0.046432 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#6	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Manipulation: · Add one or more tiles from rack to make new set: · Splitting a run:	p.2	0.045921 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300

what the answerer read — the passages it actually had in front of it

what this block IS: the passages the ANSWERING read put in front of the answerer, in the order it read them — the other half of the two reads described above, and the one that decides what the answer could possibly say. Each row names the book it came from and where in that book it sits, what it scored when the search arms were fused, which arm carried it there, and whether the re-ranker chose it or a floor guaranteed it a slot. A row here that no passage-search row names is EXPECTED: a far wider pool, re-ranked, with floors the passage search never applies.

candidate pool: the answering read scored 6 passages, out of a ceiling of 80 set gateway-side. Both figures are reported by the gateway for THIS ask — measurements of what actually happened, not constants read out of anyone's source and printed here on faith.

floors armed for this ask: the base rulebook — armed · surviving errata corrections — not armed · one slot per book on a multi-book read — not armed. Every floor is stated on every ask, armed or not, so a floor that did not apply here is never mistaken for one this app does not know about.

arms fused for this ask: vector + full-text + the doc-priority boost arm. Stated separately from the rows because an arm that ran and carried nothing is a different fact from an arm that never ran.

read at	document	heading	page	fused rank	fused RRF score	carried by	how it got its slot
#1	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Playing The Game:	p.1	#1	0.04918 (at the 3-arm ceiling (0.04918) — every arm ranked it #1)	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#2	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Manipulation: · Add one or more tiles from rack to make new set: · Splitting a run:	p.2	#6	0.045921 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#3	RULEBOOK Rummikub — Rulebook (corpus doc 152)	The Joker: · 4 ways to clear the joker are:	p.3	#3	0.047371 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#4	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Strategy:	p.4	#2	0.048131 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#5	RULEBOOK Rummikub — Rulebook (corpus doc 152)	4 ways to clear the joker are: · Time Limit: · Incomplete Runs: · Winning: · Scoring:	p.4	#4	0.046635 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#6	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Assembly instructions: · Contents for Rummikub XP/Mini XP · Object of The Game: · Sets: · Set-up:	p.1	#5	0.046432 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	vector + full-text + the doc-priority boost arm	the re-ranker chose it

chosen chunks — what the answer actually cited

#	document	heading	page	quote chars
[1]	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Assembly instructions: · Contents for Rummikub XP/Mini XP · Object of The Game: · Sets: · Set-up:	p.1	no quote on the wire
[2]	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Playing The Game:	p.1	no quote on the wire
[3]	RULEBOOK Rummikub — Rulebook (corpus doc 152)	Manipulation: · Add one or more tiles from rack to make new set: · Splitting a run:	p.2	no quote on the wire
[4]	RULEBOOK Rummikub — Rulebook (corpus doc 152)	The Joker: · 4 ways to clear the joker are:	p.3	no quote on the wire
[5]	RULEBOOK Rummikub — Rulebook (corpus doc 152)	4 ways to clear the joker are: · Time Limit: · Incomplete Runs: · Winning: · Scoring:	p.4	no quote on the wire

generation — timing and the path taken

timing split: retrieval 82 ms · first delta 1746 ms · total 5819 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
full breakdown: queued 0 ms · retrieval 82 ms · ask 5735 ms · gateway work 5819 ms · total 5819 ms
tokens in (the prompt the answerer read): 2402
tokens out (what the answerer generated): 543
tok/sec: 134.6 tok/sec — 543 output tokens in 4.034 s of generation
live seat (read from the running child): answering — gemma4:26b think:false · num_ctx 32768 (live-verified 52s ago)
answerer host (measured by rulesage): reachable (for 22h, checked 1s ago)
answerer node (the seat proxy's own header — this probe, not this ask): node1
stream path: streamed — the gateway pushed at least one accumulated draft
watchdog: first-token watchdog: 30s per arming window (rulesage.table.first-token-watchdog-seconds), re-armed once by the passages frame — so the worst case before a reader falls back to polling is 60s. Whether it actually fired is a BROWSER-side decision that never reaches the server, so it is not recorded here. The stream path above is the server's half of that story: whether a delta was emitted at all.
num_ctx · think posture: both are the CHILD's settings, not rulesage's, and no ask reply carries them — an answer never states which context window or think posture produced it. The live-seat line above reads them from the running child instead, where the gateway can answer for itself; it describes the child NOW, not this ask.

disposition

outcome: ANSWERED — Answered
abstain rationale (n/a — this ruling answered): answered — no abstain rationale applies

setup-shape — why the checklist band did or didn't render

decision: NOT_SETUP_SHAPED — no setup-shape pattern matched this question

errata registry — what the gate above had to match against

this game registers NO errata cards (edition 74), so "no registered errata card matched" above could never have matched one

Errata scope: the gate above matched this ruling's own edition only. When an ask also reads an expansion, that expansion's own registered errata cards stay outside the match — a stated phase-1 limitation, not a failed lookup.