A release whose last post is 6 hours old gets nothing from the tail, only
fragments from neighbouring ranges: settled once unchanged for 30 minutes.
Waiting 6 hours after it was read held every history release back from
phase 2 after each reset; with the hourly sweeps' resets the prober (and so
the par2 merge) never had work.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The prober stores, per probed article that starts a file, the MD5 of its
first 16 KiB (file_hashes), and per par2 body the set's whole file list
(par2_files, par2_sets): FileDesc holds the same hash per file, whatever
the file was renamed to. releases:merge (the merger service, no NNTP)
merges every minute the releases of one group, posted within 48 hours of
a release holding the set's par2, whose files the set describes, into the
one with the most files: chunks, probe candidates, inspections, hashes and
counts move; the par2's shared name when none was readable; release_merges
keeps what went where. Obfuscated uploads whose files were each a release
of their own ("[9/20] - "14f5999a…"", hep: 13,232) become one.
Also: corpus_sweep.py counts the marked ranges without psql's command tag.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ranges marked `capture` keep their XOVER chunks in CORPUS_DIR
(corpus/<group>/<start>-<end>/, edge-bench's layout, range.json once
complete; only chunks that inflate). The prefetcher reads a stored chunk
instead of fetching it, so a reset re-indexes them without the news
server; a chunk a worker reports corrupt goes from the corpus too.
edge-bench gets --releases (release chunks without payload, with their
first subjects) and ships in the edge image; deploy/corpus_sweep.py marks
more ranges (capture), replays the corpus with the deployed build (replay)
and reports suspects and the change since the previous build (report).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From the top-100 groups:
- A hashed stem under text without a counter is the upload ("Seit 1753
Tagen - usenet4ever.info - "055631a1_b6c820bc_…-u4e.part01.rar""): the
day's uploads were one release of 3,747 files.
- A numbered volume as the extension, within the [i/M] total ("(????)
[0659/2195] - "DN.VPL PCK.1508"", "TKY.ASN PCK.691."): keyed without it.
- Scene tracks under a bare counter ("[06/14] - "02-repiet_-_dont_think_
twice_(extended_mix).mp3""): the artist's album bounded by the total.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Quoted text other than the file name stays in the remainder: ">>> CREATIVE
COMMONS NZB <<< "0190 alarm 5.0 " - File 1 of 9: "…rar"" made 12,605
files of all uploads one release.
- namesRelease accepts words joined by , & : ; \ / + ("Posting & deleting,
read 0 file "E:\VladModels\y038_Irina\185" …"): such image sets fell to
the stem rule ("P8250001.JPG"), one release over all folders.
- imageSet: a 5+ digit number left in the set is a per-image id
("sss-70556-3474182-23-10825418-pc", "Alena p 26663_…"): no set. Copy
markers may repeat ("diana-050-052[1][1]").
- unquotedName: an image after numbered text without a separator is the
last word ("<abm> … AB: part2 0573-1002 043.jpg").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- The unit-less size before the closing yEnc ('"x.mp3" - 8123 yEnc') took
the set number in 'Newstar Destiny I set 1 [1 of 60] "destiny-001-001.jpg"
yEnc' once the quotes were gone: 97 sets of 60 were one release. A number
of fewer than 4 digits after a word stays.
- unquotedName also cuts after "-> " ("… -> [0001 of 0076] ->
Exclusive-tgirls_000000-lg.jpg"): the arrow was part of the file name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- An image whose subject has no words ("##[ "diana-040-001.jpg" (01/68)")
is keyed by its set, not its name without numbers: a poster's 263 diana
sets were one release of 17,808 images.
- imageSet strips "-2048px" and ImageVenue's "_123_244lo", and reads a
spaced suffix ("she has sensual legs 172 _ 172") as one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each image was a release of its own in: "Set 008 10of63 new0464.jpg",
"202/430 princess-079-058.jpg", "( 098 /500) 002 148.jpg" (the counter
is part of the name: the set is the text before it),
"sexypattycake_0903040445[1].jpg" (a copy's marker), "092_081_2000x1500.jpg"
(a resolution) and "KP0027_197_00.jpg" (a constant suffix).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- The file total is read from the subject without the file name, as with a
quoted one: "shyla-shy-09-0195.jpg hardcore …" read as 9 of 195 made each
of 866 images a release of its own.
- imageSet: "_N" right after a number is a renumbering ("kate108lg029_28"
once an image was skipped), and a trailing camera timestamp
("kurozo-2006-7-26-15-21") numbers no set.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"051 - White leather kate054001_1_1.jpg": the index is inside 054001
(set 054, image 001) and _1_1 repeats it, but only a suffix equal to the
whole number before counted as a repeat, so the set kept 054001 and every
image was a release (Duke's threads: 2,791 releases of 3,251 files). A
suffix that the number before ends with is a repeat as well. Both
implementations; tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The set bound cut at an image name's last number, but in many threads
that's a suffix, so every image was a set ("It does my heart good to see
Red Dog dancing again": 4,683 releases of 4,730 files): a repeat of the
index (katelg010_10), a year (hgrscans1097_2009), a size
(pj02kristina_24of40_1600), an "of" total (24of40). imageSet drops
those, then cuts at the last number (cutie-071-032 stays cutie-071-).
Both implementations; tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Without a file counter a subject-named image release was bounded by the
image's name with all numbers removed:
- STIRILING SILVER's sets stp250x…stp465x became one 44k-image release
(all "stp x"); picture threads of many sets (keira_s04, keira_s05,
perla_s01, apneatic-7, -8 …) one release each.
- Free-text names without numbers ("JimAndJoey Kindly Husband-to-Be.JPG")
made each image a set: one thread was 1,177 releases.
The bound is now the name up to its last number with its digits
("stp250x", "keira_s05_l keira_s05_", "apneatic-7-"), and nothing for a
name without numbers. models fixture: 172 -> 754 releases, by set. Both
implementations; tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"Avira AntiVir Rescue System 08-29-2011.rar", "Android Apps and Games
Pack 27-02-2011", "…-cable-18-08-2012-1king.rar": counterTotal read the
date as a counter (8 of 29, 2 of 2011), so the release showed a wrong
file total and counted as incomplete. Dates (2011-08-29, 08-29-2011,
29.08.2011, 02-2011) are taken out before the written-out counters are
read; [i/M] and (i/M) come first as before. Both implementations; tests;
fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"Spotlight - 83xx's [16/19] - \"SC8367-12 - Wynette, Tammy - Singing My
Song.zip\"": the disc-and-track code isn't a bare track number, so the
poster-tag rule keyed every track by its own name. A stem starting with
an upper-case code and digits before the track number (SC8367-12,
ZMP050-04) is a numbered item too. Both implementations; tests; fixtures
identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From the re-index's single-file releases:
- A quoted text without an extension ("My \"Official\" remaining set's
of Inna … 025.jpg [25/74]", "… lr140x16.jpg ( file 046 of 922)
\"yEnc\"", "\"Dr. Lust\" - … - [12/21] - Ed_DRL_VDR_LM.part09.rar") was
the file name, the real one stayed in the subject: every file a release.
It gives way to a file name with an extension beside it
(nameBesideQuotes); random quoted names with none beside keep theirs.
- Split parts numbered with two digits ("Admiral.EP05.2008.720p.Rus.
HDCLUB.mkv.21", a.b.movies.divx.russian) share their stem, as .001.
Both implementations; tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Releases of one file whose counter announces many (910k of 3M after the
sixth reset's first minutes), by cause:
- An unquoted name with double spaces ("lara_bath 15of19 larabath_2.5.jpg",
"Luisa: (068 of 160)"): the subject is collapsed, the name wasn't,
so it stayed in the remainder; it's taken out collapsed too.
- Audiobook and album tracks keyed by their own name ("RP - File 268 of
295 - \"… Seeing A Large Cat - D11.16-22.mp3\"", "CANDYD :: \"Candy
Dulfer - Candy Store - 11 - everytime.mp3\""): by the name without its
numbers, as image sets.
- A size without a unit before the closing yEnc ("…mp3\" - 8123 yEnc").
- Random posters with separators ("SOFZRw1-OTEd@…", "…_OOW@…").
Both implementations; tests; fixtures identical (models: 267 -> 172).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"A Strange Light from the East - Eastern Influenced Western Music
(1965-72) - CD2 - 07 - Rascals - Sattva.mp3 (1/43)": unquotedName cut at
the last " - " ("Sattva.mp3"), leaving track number and artist in the
subject, so every track was a release (85665085, 85665050). With a
" - NN - " track or part number the name starts there ("07 - Rascals -
Sattva.mp3"), the rest names the album and disc. Both implementations;
tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
FTD posts one upload's files to different groups ("Knowsum-Loops_Are_My_
Religion-WEB-2026-FTD": par2 and part1 to a.b.music.misc, part2 to
a.b.sounds.music, each article in one group only; releases 81825221 and
81825412): the catalog matched a chunk only to releases of its own
group. A chunk without one there now joins a release of its key in
another group whose window is within 3 hours, when none of its file
names are in that release yet; a post of the same files to another group
(other message-ids, same names) stays a release of its own, or its NZB
would hold every file twice. Functional tests for both.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Releases 78361820 and 78411563 were two files of one FTD upload
("Yusef_Lateef-Alight_Upon_the_Lake_…-WEB-2026-FTD", [1/6] and [3/6], 4 s
apart) under a new random poster each ("N9vplIaavBCz@aCTVoAsk.JSE",
"FVWWHku0fGAg@QfaTz0g0.RVH"): the stem key includes the poster. When the
poster's own name (before the @) looks random, it's left out of the key;
real posters keep theirs (two uploaders of one scene release at the same
time stay apart), and uploads hours apart stay apart by the re-upload
rule. Both implementations; tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Old picture groups split every image into a release (the reset's first
hour: 583k releases, 447k of one file; a.b.pictures.erotica.high-heels
262k): "Aletta Ocean - \"100628-035.jpg\" yEnc [347K]" kept its size in
the remainder, and "Corie - \"2877-041.jpg\" yEnc [290K]" named its set
with one word, which namesRelease (two words) refused, so the image's
own name keyed it. Sizes like [347K] (and a cut-off "336573 byt") are
removed; for images a one-word subject (3+ letters) names the set. On
100k high-heels articles: 4,509 releases -> 192. Both implementations;
tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"ai-engineer-complete [401/465] - \"003 Defining Tools with JSON
Schema.mp4\"": the subject's text shares no word with the lesson's
descriptive name, so it was taken for a poster's constant tag and each of
the course's 465 files keyed a release of its own (releases 75941915,
75941916). A stem starting with a 1-3 digit index and a separator is an
item of a set the subject names, as audio tracks and images already
were; uploads hours apart stay apart by the re-upload rule. Both
implementations; tests; fixtures identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Most "obfuscated" subjects of older and smaller groups name their file
without quotes ("DVDFab 8.2.0.5.rar (006/332)", fta's "q3B9…part023.rar
(104/137)", "New_Super_Mario_Bros._[Keygen].exe (1/2)") or with one
quote ("[keira_s04_0037_l.jpg\" yEnc] (2/2)"): 100% of ath 2013's and
fta's, 94% of models' (LOG.md 2026-10-11 (5)). With a segment counter and
no quoted name, unquotedName takes the name before a lone quote (after a
"[", else after the last space), else the text when it ends in a file
name (after "] " or " - "), else its first word that is one; a file
counter stuck in a bracket is dropped. The release identity takes such a
name out of the subject as it would quotes (else every image of a set
was a release: 9,130 instead of 267 on the models fixture).
Both implementations; a fourth fixture (fta, 25k lines) and tests; check
identical on all four. Truly random subjects (ngPost hex = message-id,
random tokens) stay obfuscated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
obf_runs: how obfuscated articles come in runs (strict, and per poster
and tool with a time gap). obf_kinds: what they are (unquoted subjects
with a counter, random tokens equal to the message-id, other random
tokens, ...), with examples. LOG.md 2026-10-11 (5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
User: Rust is the product (option a). tools/edge-bench-rs moves to edge/
(crate edge; binaries edge-node, the worker, and edge-bench, the replay).
CI's new job edge runs cargo test and edge/check-fixtures.sh: both
RangeIndexers on three committed captures (edge/fixtures, 25k overview
lines each of a.b.models, a.b.ath and a.b.boneless in 5 chunks, 3.5 MB),
compared field by field with check.ts; deploy waits for it. The
TypeScript RangeIndexer stays as check.ts's reference. README: what it
is now; the benchmark findings as history.
Random names: looksRandom missed long tokens of capitals and digits
("51294P1CUKR2R35X", release 74539068, readable and a Newznab title).
12+ capitals and digits switching 6+ times between them are random
(COD4MW2REMASTERED switches 4 times, BEATLES1964REMASTER twice); both
implementations, identical on the fixtures.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The node API (leases, hour-long streams, batches of up to 256 MB JSON
parsed in-process) shared the web process with the catalog API, Newznab
and the web interface; voyager 1 had to make the same split on
2026-10-06. HTTP_ROUTES picks a process's routes (node, public, all by
default): compose service node-api (same image, port 3335 on the compose
network) serves the workers, web the rest. The workers' COORDINATOR_URL
points at node-api; deploy.sh stops edge before recreating node-api.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From "draft for approval" to the concept of voyager 2 (voyager 1's
CONCEPT.md and D1-D42 are superseded). Phase 2: what's built (the
prober), and voyager 1's features not yet here: spots, NZB import,
passwords, SFV, RAR walks, 7z listings, body naming of obfuscated
subjects, after-probe garbage drops; where its exported data is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The default page (no search words) took 23-27 s: SearchService sorts
posted_at DESC NULLS LAST, id DESC, the indexes were on posted_at DESC
(nulls first), so Postgres sorted all 68M releases. New indexes match
the sort (also per category), built concurrently; the old ones dropped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Search matches a word in the name or the inner name (trigram index on
inner_name, partial, built concurrently); "Readable names" includes
releases with an inner name.
- unlocked=1 (chip "No password") leaves out passworded releases.
- Results carry passworded and innerName: a Password badge, the inner
name under the release name.
- The release view: Inside and Inspected facts, a passworded notice, and
Contents (per probed article its yEnc name, kind, type, encryption and
the names inside) from release_inspections.
- Newznab: the password attribute from the prober; an obfuscated release
with an inner name is included, titled by it (without its extension).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Postgres evaluated random() once for all 20 slices: every claim still
came from one place in the queue (214 of 214 taken from a.b.zogg).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
In id order the prober went group by group, and some groups' articles
take 0.6-3 s each (a.b.zogg 2016; repeated fetches the same, pipelining
4 deep no faster), where a sample across all years takes 20-60 ms for
most: 7 articles/s on 6 connections, against 67/s in its first minutes.
A claim now takes 20 slices from random points (353 ms per 200).
tools/body_timing.ts: BODY time for the oldest and newest candidates.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Lone candidates are often par2 volumes of random-named sets
("42507096….vol007+08.par2"): their yEnc name was taken as the inner
name. Packaging names are skipped for lone files too, and a name that
starts with the release's own name says nothing new.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A release can have several candidates of a kind (two RAR sets): keyed by
(release, kind) both landed in one upsert ("ON CONFLICT DO UPDATE
command cannot affect row a second time"), and the batch was lost.
release_inspections is keyed by candidate now; the first run's results
are cleared and probed again. A random-named file's yEnc name is often
its random name: an inner name equal to the release's name is dropped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
probe:run (compose service prober, PROBE_CONNECTIONS, default 6) claims
waiting candidates of settled releases (lone, then par2, then archive),
fetches each article with BODY over connections of its own (no
compression) and inspects it: the yEnc name and size, the content type
from the first bytes, a par2 set's files, a RAR volume's entries and
whether they're encrypted. Results per release and kind in
release_inspections (CHUNK_TABLESPACE, the HDD); on the release
passworded, inner_name (the largest content name) and inspected_at.
430 = missing, not yEnc = error, both final; connection trouble puts
the candidate back.
Fetch and parse in one process: an article takes ~0.1 s per connection,
parsing a few ms (no queue in between). Parsers from voyager 1 (lib/yenc,
lib/rar, lib/par2, lib/content sniff) with their tests; XoverClient.body;
yencode.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Range 161820 failed with "doesn't inflate, also fetched plain" though
the chunk was likely the compressed copy again: in flight, or spooled
before the prefetcher's targets were refreshed, when the plain flag was
set. The range now fails only on the third report.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ranges with a chunk or two left take fewer than perRange connections:
with loops/perRange + 2 ranges, 3.2 of 40 loops were idle on average
over 15 minutes (up to 9.5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Writers lock their group only, so two groups' batches can number the
same identity key at once (a spam post crossposted to 67 groups:
duplicate key ...:65, HTTP 500, the batch retried). Releases are now
inserted with ON CONFLICT (release_key) DO NOTHING, and those skipped
take the next number (up to 20 tries).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Range 154218 (a.b.hdtv.tv-episodes) looped 13 times on HTTP 413: a
500k-article batch over the server's 256 MB body limit, though under
100k files. On a 413 the worker gives the range back and posts that
range's batches at half the articles and files from then on (again on
another 413, down to 1/64); stored batches stay. The message says how
large the batch was.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Some groups' chunks take 20+ minutes while still coming in (range
154153: 17 MB in 22 minutes); the stream's 15-minute wait failed their
workers (two in 15 minutes). Stalled requests are given up separately.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a.b.0day.stuffz sends steadily but slowly (~400 kB a minute, 5-7
minutes per 100k chunk): the fixed deadline (90 s, doubling up to 360 s)
gave its chunks up again and again, and two of its ranges' worker
streams timed out. A stall is a request that got less than
NNTP_STALL_KB_PER_MINUTE (100) in the last minute (the stuck replies
got ~25 kB a minute); then its connection is closed and the chunk
fetched again. Replaces NNTP_CHUNK_DEADLINE_SECONDS.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
At 90 s a chunk that always takes longer would be given up forever
(range 155437's worker gave up waiting on its stream). Each deadline a
chunk hits doubles its next one (90, 180, 360 ... up to 16x), reset
once it comes in.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
At 180 s, 60 chunks were given up in 15 minutes: each held a connection
3 minutes, ~30% of the 40 connections' time, while finished chunks
took 24 s on average (Eweka slow this afternoon).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A file name with a NUL ("…Brutal Erpresst.ra\u0000…") failed every
batch of range 155595 (HTTP 500, jsonb: \u0000 cannot be converted to
text), so the range looped. fromWire drops NUL from every string.
The chunk deadline was given up 258 times in 3 hours at 300 s (~1.4 a
minute, each holding a connection for 5 minutes): 180 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With 2 connections per range the prefetcher needs 20+ ranges with
chunks left; fully spooled ranges counted as ready (SPOOL_AHEAD_PER_LEASE)
but take no connection, so on small ranges 30 of 40 connections were
busy (13-15 ranges). Fresh ranges now also fill up to loops/perRange + 2
ranges being fetched.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A worker waited on one chunk of a leased range for an hour (range
146477: five 15-minute stream timeouts, then the range took 63 s; 88
"stream: Peer disconnected" overnight). The silence timeout doesn't
catch a reply that trickles: a chunk still coming in after
NNTP_CHUNK_DEADLINE_SECONDS (300) is given up, its connection closed,
and fetched again. The report names the oldest request in flight (range,
chunk, age, bytes in) and counts deadlines.
Batches failed with "Maximum call stack size exceeded" (8 overnight):
Math.min(...segments) on a fragment with more segments than the stack
takes arguments.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Dockerfile with two targets: app (compiled AdonisJS, production
dependencies, Node 24.16 as in CI) and edge (the Rust worker, cargo
caches as build mounts). deploy/compose.yaml, project usenet-rebuild:
postgres (data on the HDD, catalog tablespace on the NVMe), migrate,
web, coordinator, edge; edge-ts and ace behind profiles.
deploy/deploy.sh (~/bin/deploy-usenet-rebuild on homelab): fetches the
commit (must be on rebuild), builds, migrates, recreates web,
coordinator and edge (the worker stopped first), never postgres.
CI deploys a push to rebuild that passes: the deploy key's forced
command passes "deploy-rebuild <sha>" on. A push to master no longer
deploys from this branch's workflow. deploy/README.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
At most PREFETCH_PER_RANGE (2) connections on a range and
PREFETCH_PER_GROUP (4) on a group; SPOOL_AHEAD_PER_LEASE 2 -> 4 so there
are targets enough. The report says how many ranges and groups are in
flight.
Homelab, 40 connections, alternated 12 min each: unlimited (~10 ranges
in flight) 281k and 167k articles/s; spread (~22 ranges) 319k and 296k,
chunks 11.5 s against 13-24 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Once a second: pool connections with a command in flight, loops
fetching/writing/idle; per request GROUP time, XOVER time to the status
line and the transfer, transfers with a silence over 5 s. The old
"connections busy" (fetch time of the chunks finished in the minute)
said 59-75% while all 40 connections were in flight all the time:
dropped.
Homelab: 39-40 of 40 in flight, no idle loops; the status line comes at
once, the whole 8-35 s is the transfer, with data flowing (few gaps
over 5 s): ~3 MB per chunk at 0.1-0.4 MB/s per connection.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
