frank-small: seam follows the f16 K + q8_0 V contract from 1862dc3

1862dc3 repriced the Ornith rung (nine full-attn layers: f16/f16 = dead
boot) and flipped CACHE_TYPE_V to q8_0, but its note that the seam was
'retargeted in 8f53df7' covered only the payload paths — the KV test
still asserted the 2026-10-03 unquantized-both-halves contract and held
the suite red. Test now pins the shipped contract: K=f16, V=q8_0,
flash-attn on.
This commit is contained in:
Frank committed 2026-10-05 18:23:44 -07:00
1 parent 1c725ea024
commit 5573ca171f
1 file changed
+8 -5
+8 -5
View File
@@ -118,14 +118,17 @@ argv_value() {
[ $((ctx / par)) -le 262144 ] [ $((ctx / par)) -le 262144 ]
} }
@test "the KV contract is unquantized both halves, flash-attn on" { @test "the KV contract is f16 K + q8_0 V, flash-attn on" {
serve_argv serve_argv
[ "$status" -eq 0 ] [ "$status" -eq 0 ]
[ "$(argv_value --cache-type-k)" = "f16" ] [ "$(argv_value --cache-type-k)" = "f16" ]
# Frank, 2026-10-03: the values are unquantized too. The flag pair is what # Frank, 2026-10-05 (1862dc3): the Ornith payload has NINE full-attention
# makes this observable — a build without the FA path silently falls back, # layers (18 KiB/token/half) and +1,900 MiB of weights vs MiMo Q6_K, so
# so both must appear in the argv the supervisor runs. # f16/f16 lands ~379 MiB free — a dead boot; q8_0 V keeps 2 x 200,704
[ "$(argv_value --cache-type-v)" = "f16" ] # AND the 2-3 GiB floor. The flag pair is what makes this observable — a
# build without the FA path silently falls back, so both must appear in
# the argv the supervisor runs.
[ "$(argv_value --cache-type-v)" = "q8_0" ]
[ "$(argv_value --flash-attn)" = "on" ] [ "$(argv_value --flash-attn)" = "on" ]
} }