captchakraken 2.8.0 → 2.9.0

package/dist/trajectory.d.ts

@@ -18,7 +18,7 @@
  *
  * Camoufox's own `humanize` is a DIFFERENT, browser-level mechanism that
  * re-humanises every `mouse.move()` it is handed. Running both composes them —
- * measured 82.1s vs 13.4s on one geetest_v4_slide solve, because each of the 60
+ * measured 82.1s vs 13.4s on one a GeeTest v4 slider solve, because each of the 60
  * points below became its own humanised sub-trajectory. Drive with humanize off.
  */
 export type Point = [number, number];

package/dist/trajectory.js

@@ -19,7 +19,7 @@
  *
  * Camoufox's own `humanize` is a DIFFERENT, browser-level mechanism that
  * re-humanises every `mouse.move()` it is handed. Running both composes them —
- * measured 82.1s vs 13.4s on one geetest_v4_slide solve, because each of the 60
+ * measured 82.1s vs 13.4s on one a GeeTest v4 slider solve, because each of the 60
  * points below became its own humanised sub-trajectory. Drive with humanize off.
  */
 Object.defineProperty(exports, "__esModule", { value: true });

package/dist/types.d.ts

@@ -407,6 +407,18 @@ export interface CaptchaKrakenConfig {
      */
     keyframeWaitTimeoutMs?: number;
     keyframeWaitPollMs?: number;
+    /**
+     * Hard ceiling on a burst that never repeats a screen. The burst normally
+     * ends when the cycle closes; this bounds a continuous animation, which never
+     * closes one. Default 12000ms.
+     */
+    videoBurstMaxMs?: number;
+    /**
+     * Record while the still screenshot is being read, so a cycling board is
+     * known before its answer is acted on. Default on. Off restores the older
+     * shape, which learned the same thing two rounds later.
+     */
+    speculativeBurstEnabled?: boolean;
     /**
      * Extra wall clock (ms) granted ONCE, the first time a solve escalates to a
      * recording. NOT a looser `overallSolveTimeoutMs`.
@@ -420,8 +432,8 @@ export interface CaptchaKrakenConfig {
      * timeout — which reads as a slow model rather than as a budget with no room
      * for what the solver had just decided to do.
      *
-     * Measured 2026-08-22, Tier 3 run 32596340560: hcaptcha_click_image_by_traits,
-     * hcaptcha_connect_path and hcaptcha_grid_3x3_property each failed this port
+     * Measured 2026-08-22, a full driver-gate run: an hCaptcha click-by-trait board,
+     * an hCaptcha connect-the-path board and an hCaptcha 3x3 property grid each failed this port
      * with "Captcha solve timed out after 45000ms (attempt 6/6)" at 49-59s, and
      * solve in 14-20s on the rounds the still path answers them.
      *

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "captchakraken",
-  "version": "2.8.0",
+  "version": "2.9.0",
   "description": "Captcha solver for Playwright and Puppeteer. Finds the captcha, reads the puzzle with a fine-tuned Qwen3.5-9B vision model, and clicks, drags, slides or types through to a token — reCAPTCHA, hCaptcha, GeeTest, NetEase Yidun, Tencent, Lemin, Prosopo, Turnstile and distorted text. Hosted API or self-hosted.",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -13,6 +13,7 @@
   "scripts": {
     "prebuild": "node scripts/copy-python.mjs",
     "build": "tsc",
+    "demo": "tsx examples/demoSites.ts",
     "typecheck": "tsc --noEmit -p tsconfig.json",
     "test": "tsc -p tsconfig.test.json && node --test .test-build/*.test.js",
     "prepublishOnly": "npm run build",

package/python/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "captchakraken"
-version = "2.8.0"
+version = "2.9.0"
 description = "Captcha-solving engine and CLI: OpenCV grid detection plus a fine-tuned Qwen3.5-9B vision model, for image grids, click and drag puzzles, sliders, distorted text and animated challenges. Hosted API or self-hosted on vLLM."
 readme = "README.md"
 requires-python = ">=3.10"

package/python/src/captchakraken/__init__.py

@@ -107,4 +107,4 @@ __all__ = [
 # It said 2.6.0 while the wheel on PyPI said 2.6.1, so anyone gating on the
 # runtime attribute saw a version that had not been current for weeks.
 # tests/test_public_contract.py compares all three.
-__version__ = "2.8.0"
+__version__ = "2.9.0"

package/python/src/captchakraken/cli.py

@@ -542,6 +542,10 @@ def _handle_solve_animated() -> bool:
             # solve log rather than only inferable from the answer.
             "keyframes": paths,
             "keyframe_mode": kfset.mode,
+            # What the board LOOKS like, as distinct from how it was
+            # sliced. The driver gates its clicks on this; see
+            # KeyframeSet.steady_screens.
+            "steady_screens": kfset.steady_screens,
             "source_frames": len(imgs),
         }))
     except UnsupportedCaptchaError as e:

package/python/src/captchakraken/humanize.py

@@ -14,7 +14,7 @@ three directions at once —
     device farm, a model of their own users) was composing two of them. That is
     not hypothetical — camoufox's `humanize` juggler re-humanises every
     `mouse.move()` it is handed, and running both measured 82.1s against 13.4s
-    on one geetest_v4_slide solve, because each of the 60 trajectory points
+    on one a GeeTest v4 slider solve, because each of the 60 trajectory points
     became its own humanised sub-trajectory.
   - a caller on their own infrastructure, against their own fixtures, is paying
     for texture nobody is measuring.
@@ -41,7 +41,7 @@ pointer", because that is what the next gesture starts from and what
 `SolveResult.final_mouse_position` reports.
 
 The TypeScript side is `js/src/humanize.ts`; the two are one design in two
-languages, and `tests/test_humanizer_parity.py` in CaptchaKrakenFinetune pins
+languages, and `the training repo's humanizer parity tests` in the training repo pins
 that the mode names and the pause vocabulary have not drifted.
 """
 

package/python/src/captchakraken/keyframes.py

@@ -1,8 +1,8 @@
 """Reduce a recorded captcha clip to the few frames the model is shown.
 
 VERBATIM PORT. Everything below the docstring is byte-identical to
-`src/video/keyframes.py` in the CaptchaKrakenFinetune repo, and
-`scripts/check_prompt_parity.py` there fails the build if the two ever diverge.
+the training-side extractor in the training repo, and
+`scripts/the parity gate` there fails the build if the two ever diverge.
 
 That is not tidiness, it is a correctness requirement. The model is trained on
 keyframes cut by that code and answers with a frame NUMBER indexing into them. If
@@ -41,7 +41,7 @@ DEFAULT_MAX_KEYFRAMES = 6
 # picture". A false "same" merges two distinct states, which loses an answer, so
 # this is the most consequential number in the file.
 #
-# MEASURED, not guessed. Over all 20 real clips in cleanSamples/test/raw, the
+# MEASURED, not guessed. Over all 20 real clips in the sample corpus, the
 # consecutive-frame differences fall into two clusters with a 64x gap:
 #
 #   same picture (compression + antialiasing)  <= 0.000067   (0.067 per mille)
@@ -162,6 +162,15 @@ class KeyframeSet:
     # `keyframes`). Empty for the other modes. Lets a caller answer "which
     # keyframe was the clip showing at source frame i?" without re-diffing.
     frame_states: List[int] = field(default_factory=list)
+    # How many distinct steady screens the clip sits on; 0 if it is not that
+    # kind of clip. INDEPENDENT OF `mode`, and the difference is the whole
+    # point: `mode` says what the slicer could PROVE about recurrence from one
+    # burst, this says what the board looks like. A three-screen captcha whose
+    # loop is longer than the recording is `even` with `steady_screens == 3`.
+    # The live driver reads this to decide whether waiting for a named screen
+    # is meaningful; reading `mode` for that put the wait off on 100% of real
+    # animated captchas. See `steady_screens()`.
+    steady_screens: int = 0
 
     def __len__(self) -> int:
         return len(self.keyframes)
@@ -226,6 +235,7 @@ class KeyframeSet:
         return {
             "stem": stem,
             "mode": self.mode,
+            "steady_screens": self.steady_screens,
             "fps": self.fps,
             "source_frames": self.source_frames,
             "params": self.params.as_dict(),
@@ -440,14 +450,22 @@ def _distinct_indices(
     return sorted(kept)
 
 
-def _detect_cycle(
+def _steady_screens(
     frames: Sequence[np.ndarray], params: KeyframeParams
-) -> Optional[Tuple[List[int], List[int]]]:
-    """`(state_medoid_indices, per_frame_state)` if the clip is a cycle of steady
-    pictures, else None.
+) -> Optional[Tuple[List[Tuple[int, int]], List[int], List[int]]]:
+    """`(holds, state_medoid_indices, state_per_hold)` if the clip is a set of
+    steady held pictures, else None.
 
-    `per_frame_state[i]` is the 0-based state of source frame i, or -1 for a
-    transition frame that belongs to no hold.
+    Split out of `_detect_cycle` because the answer is wanted TWICE and the two
+    questions are not the same one:
+
+      - "is this a cycle?" — needs a state to be seen coming BACK, which takes
+        more than one pass through the loop.
+      - "does this board sit on a few stable screens?" — needs only one pass,
+        and is what the live driver has to know in order to decide whether
+        waiting for a particular screen is meaningful.
+
+    Answering only the first left the second unasked. See `steady_screens`.
     """
     n = len(frames)
     runs = _anchor_runs(frames, params.steady_ratio)
@@ -479,6 +497,55 @@ def _detect_cycle(
     if len(state_reps) < 1:
         return None
 
+    # NO `distinct_ratio` HERE, deliberately. The merge above already separated
+    # these states by more than `steady_ratio`, which is the measured noise
+    # floor — that is the whole of "are these different pictures".
+    #
+    # `distinct_ratio` (0.02) is twenty times coarser and exists to license
+    # `cycle`'s COLLAPSE to one frame per state, where being too generous throws
+    # away the middle of a clip. Applying it here instead answers a question
+    # nobody asked and answers it wrongly: measured on a live GeeTest svg burst,
+    # the three screens are separated by 0.0056 and 0.0050 — thin line art, a
+    # few glyph strokes change — so a 0.02 bar reports a board with three
+    # obvious screens as having none. keyframes.py's own comment already said
+    # so ("real state separations top out around 0.007, well under this") and
+    # read it as evidence the check was harmless.
+    return holds, state_reps, hold_state
+
+
+def steady_screens(frames: Sequence[np.ndarray], params: KeyframeParams) -> int:
+    """How many distinct steady screens this clip sits on. 0 if it does not.
+
+    THE LIVE DRIVER'S QUESTION, and it is not `mode`. A board that advances
+    through three held pictures gives a burst that contains ONE pass, so
+    `_detect_cycle` cannot prove it recurs and the clip is sliced `even` — which
+    is right for slicing (dedup reaches the same stills) and was being read by
+    the driver as "nothing recurs, do not wait for a screen". Every real
+    geetest_v4_svg clip is exactly this shape: 2-3 holds covering 95-100% of the
+    burst, mutually distinct, and the page absolutely does come back around.
+
+    A one-way animation — the case the no-wait path exists for — is nearly all
+    transition, so it has no qualifying holds and this returns 0.
+    """
+    got = _steady_screens(frames, params)
+    return len(got[1]) if got else 0
+
+
+def _detect_cycle(
+    frames: Sequence[np.ndarray], params: KeyframeParams
+) -> Optional[Tuple[List[int], List[int]]]:
+    """`(state_medoid_indices, per_frame_state)` if the clip is a cycle of steady
+    pictures, else None.
+
+    `per_frame_state[i]` is the 0-based state of source frame i, or -1 for a
+    transition frame that belongs to no hold.
+    """
+    n = len(frames)
+    got = _steady_screens(frames, params)
+    if got is None:
+        return None
+    holds, state_reps, hold_state = got
+
     # A cycle REVISITS. More holds than states means some picture came back, which
     # is what "the video is just a shift between N steady images" describes. Equal
     # counts mean every hold was a new picture — a one-way progression (a slow
@@ -497,8 +564,10 @@ def _detect_cycle(
     if len(holds) <= len(state_reps) and len(state_reps) > 1:
         return None
 
-    # Every pair of states must be clearly different, or we are calling noise a
-    # state and would hand the model two images of the same thing.
+    # Every pair of states must be clearly different before the clip is
+    # COLLAPSED to one frame per state — that collapse is what this threshold
+    # licenses, and being too generous with it discards the middle of a clip.
+    # `steady_screens` deliberately does not apply it; see `_steady_screens`.
     for a in range(len(state_reps)):
         for b in range(a + 1, len(state_reps)):
             if frame_diff_ratio(frames[state_reps[a]], frames[state_reps[b]]) < params.distinct_ratio:
@@ -538,7 +607,7 @@ def _drop_smeared(
         six eval clips, dropping a SCREEN. A page that never reaches the model is
         unanswerable whenever the target lived on it.
       * *Drop when some other still is within `distinct_ratio`* cut
-        `hcaptcha_tile_flip_video` from 6 stills to 3, because two boards of that
+        an hCaptcha tile-flip animation from 6 stills to 3, because two boards of that
         puzzle differ by ONE TILE — well inside `distinct_ratio` — so a diff
         threshold cannot tell "the same picture again" from "the next board".
         EXPECTED_SLICING says it directly: fewer than 6 "means dedup merged two
@@ -625,6 +694,7 @@ def extract_keyframes(
         reps, per_frame = cycle
         mode = "static" if len(reps) == 1 else "cycle"
         return KeyframeSet(
+            steady_screens=len(reps),
             mode=mode,
             keyframes=[
                 Keyframe(number=k + 1, source_index=idx, timestamp_ms=_ms(idx),
@@ -654,6 +724,9 @@ def extract_keyframes(
         # smears the first pass had just removed.
         indices = _drop_smeared(frames, indices, p)
     return KeyframeSet(
+        # Asked even though the cycle test refused: the two questions differ,
+        # and this is the one the driver needs answered.
+        steady_screens=steady_screens(frames, p),
         mode="even",
         keyframes=[
             Keyframe(number=k + 1, source_index=idx, timestamp_ms=_ms(idx),
@@ -679,7 +752,7 @@ def extract_keyframes_from_video(
 #   <media_dir>/keyframes/<stem>/keyframes.json
 #
 # Derived, regenerable, and ADDITIVE — the clip stays canonical next to it. That
-# matters most under `cleanSamples/test/`, which is irreplaceable hand-labeled
+# matters most under the held-out sample corpus, which is irreplaceable hand-labeled
 # data we never rewrite (CLAUDE.md § Data trees): extraction only ever creates a
 # new sibling directory there.
 

package/python/src/captchakraken/models.json

@@ -62,7 +62,7 @@
     "               authorised holder can do: a licensed model has nothing to",
     "               download however good your credentials are. `latest` may",
     "               never be private either — an unpinned self-hoster would get",
-    "               a 401 instead of a model — and check_prompt_parity.py says so.",
+    "               a 401 instead of a model — and the release parity gate says so.",
     "",
     "  \"licensed\" — NOT downloadable by anyone. Reachable only through",
     "               api.captchakraken.com, or by a holder of a licence we issue.",
@@ -73,8 +73,8 @@
     "place that decides what an unpinned client downloads and what prompts it",
     "sends. A second list of 'models you may not publish' kept somewhere else",
     "would be a list that drifts, and the drift is silent in the direction that",
-    "costs the most. The finetune repo's publisher reads this field and REFUSES",
-    "to upload a licensed model to the Hub; check_prompt_parity.py refuses to let",
+    "costs the most. The publisher reads this field and REFUSES",
+    "to upload a licensed model to the Hub; the release parity gate refuses to let",
     "`latest` be one, because an unpinned self-hoster would resolve to weights",
     "they cannot obtain.",
     "",
@@ -102,7 +102,7 @@
     "",
     "Each expert name must resolve — through `served_aliases` or as a repo id —",
     "to a REGISTERED entry declaring the same `prompt_version` and the same",
-    "`pixel_budget` as the model that routes to it. check_prompt_parity.py",
+    "`pixel_budget` as the model that routes to it. The release parity gate",
     "enforces that. It is what makes aliasing four arms onto one entry safe",
     "rather than convenient: the four share a base, a schema and a band today",
     "(measured, not assumed — runs 20260906-0142xx all carry prompt generation",
@@ -132,7 +132,7 @@
         "min": 518400,
         "max": 518400
       },
-      "note": "PUBLIC on the Hub under Source-Available License v1.1, and the first published generation-2 model. TRAINED with floor 448^2 / ceiling 720^2, but MEASURED best served at a flat 720^2 — floor == ceiling, so every image arrives at exactly one size and the band cannot drift. The band was chosen by sweeping the serving floor over 448/576/640/704/720/736/768/896/1024^2 and scoring each arm: the plateau is 704-736^2, and it falls off above 768^2, where a 400x580 portrait capture runs more than 40% past the ViT's 48x48 position grid. Do not \"just use the ViT's native grid\": 640^2 is where that grid is exactly saturated for a portrait capture, and it scores WORSE than 720^2 — mild position-embedding extrapolation plus real pixels beats staying inside the grid. Was served as `captcha-v12` while absent from this file, which resolved it to generation 1 prompts against generation-2 weights — the failure this file exists to prevent."
+      "note": "Generation 2, and the first published one. Trained over a range of input sizes but served at a flat 720^2: floor == ceiling, so every image arrives at exactly one size and the band cannot drift. The band was chosen by sweeping the serving floor and scoring each setting; the plateau is 704-736^2 and it falls off above 768^2, where a portrait capture runs well past the vision tower's position grid. Note that the vision tower's native grid is NOT the best setting — mild position-embedding extrapolation over real pixels beats staying inside the grid."
     },
     "CaptchaKraken/Twilight-v1.2-FP8": {
       "prompt_version": "2",
@@ -142,7 +142,7 @@
         "min": 518400,
         "max": 518400
       },
-      "note": "CaptchaKraken-Lora-v1.2 merged into Qwen/Qwen3.5-9B at FP8-dynamic. The weights the hosted API answers with. Generation 2, same band as the adapter it is made from: a merge that resolved to a different prompt generation or a different pixel budget than its own LoRA would be the 2026-07-18 regression with extra steps."
+      "note": "CaptchaKraken-Lora-v1.2 merged into Qwen/Qwen3.5-9B at FP8-dynamic. Generation 2, same pixel band as the adapter it is made from — a merge that resolved to a different prompt generation or a different band than its own LoRA is exactly the mispairing this file exists to prevent."
     },
     "CaptchaKraken/Sunlight-v1.2-AWQ-4bit": {
       "prompt_version": "2",
@@ -160,7 +160,7 @@
       "base_model": "RedHatAI/Qwen3.5-9B-FP8-dynamic",
       "lora_revision": "main",
       "lora_name": "captcha",
-      "note": "Currently served in production. PRIVATE on the Hub, so prompts.json can never be fetched for it — this entry is the only thing that maps it to a prompt generation."
+      "note": "Generation 1. Private on the Hub, so prompts.json cannot be fetched for it — this entry is the only thing that maps it to a prompt generation."
     },
     "CaptchaKraken/CaptchaKrakenV1_Lora": {
       "prompt_version": "1",
@@ -168,19 +168,19 @@
       "base_model": "RedHatAI/Qwen3.5-9B-FP8-dynamic",
       "lora_revision": "main",
       "lora_name": "captcha",
-      "note": "Published before prompts.json existed — no such file in the repo, so it resolves from here."
+      "note": "Generation-1 LoRA on Qwen3.5-9B. Superseded by v1.2; kept registered so an existing pin keeps resolving to the prompts it was trained on."
     },
     "CaptchaKraken/Sunlight-AWQ-4bit": {
       "prompt_version": "1",
       "kind": "merged",
       "lora_name": "captcha",
-      "note": "Ships its own prompts.json; this entry keeps it resolvable offline."
+      "note": "Generation 1 merged into the base model at AWQ 4-bit."
     },
     "CaptchaKraken/Twilight-FP8": {
       "prompt_version": "1",
       "kind": "merged",
       "lora_name": "captcha",
-      "note": "Ships its own prompts.json; this entry keeps it resolvable offline."
+      "note": "Generation 1 merged into the base model at FP8-dynamic."
     },
     "CaptchaKraken/Abyss": {
       "availability": "licensed",
@@ -198,7 +198,7 @@
         "video": "abyss-video",
         "text": "abyss-text"
       },
-      "note": "IN DEVELOPMENT. Never published, never downloadable — see `availability`. Registered before it exists so the licence gates are armed rather than retrofitted after the first artifact is sitting on a disk: the publisher, the parity gate and `captchakraken fetch` all read this entry, and a rule added the day it is first needed is a rule that was absent on the day it was first needed. Being nameable is not being downloadable: `availability` is what refuses the weights, and the two are independent on purpose — a licence holder DOES name this model, which is why it and each of its experts has a served_aliases entry. ABYSS IS THE ROUTED 9B, NOT THE 27B. It was going to be a single Qwen3.8-27B adapter; that model was trained (run 20260903-235101) and did not earn its cost, so the product became the four-expert mixture on Qwen3.5-9B that docs/MOE_LORA_DESIGN.md §7 and §10 specify. A routed 27B may be a later generation of this same name; it would be a new entry when it exists, because the base model is what a self-hoster loads. `lora_name` is `abyss-general` rather than a fifth name: the generalist IS the fallback for any family `experts` does not map, so pointing the two at one adapter is what makes §11’s “degrade to the generalist, never 400” true with one fewer resident LoRA slot — and slots are the binding constraint (§10, five names against `--max-loras`). prompt_version 2 and the flat 518400 band are what all four arms are training at TODAY (verified against training_runs/20260906-0142xx/run.json, min_pixels == max_pixels == 518400); CONFIRM BOTH from the run records at publish time and re-record here in the publishing commit, and if any arm has diverged give that arm its own entry — check_prompt_parity.py refuses the shared alias rather than resolving it silently."
+      "note": "A routed mixture of four expert LoRAs on Qwen3.5-9B — grid, pixel, video and text — selected by prompt family. Hosted only: `availability` refuses the weights, and being nameable is not being downloadable. A licence holder does name this model, which is why it and each of its arms has a served_aliases entry. `lora_name` is `abyss-general` rather than a fifth name because the generalist IS the fallback for any family `experts` does not map, so the two point at one adapter and the mixture needs one fewer resident LoRA slot."
     },
     "CaptchaKraken/Abyss-27B": {
       "availability": "private",
@@ -211,7 +211,7 @@
         "min": 518400,
         "max": 518400
       },
-      "note": "PRIVATE on the Hub. The single-adapter 27B that Abyss started as — run 20260903-235101, re-gated as 20260906-025040. NOT the routed model and not one of its arms: it is a Qwen3.8-27B, and the base model is what a self-hoster loads, so it could never share the `CaptchaKraken/Abyss` entry however similar the product story is. Kept because it is measured and it wins: Tier 2 gated_overall 0.6699 against the v1.2 baseline's 0.6456, solved 0.7113, forgetting 0.0052. What it did not win was its COST — see the `CaptchaKraken/Abyss` note — which is a statement about serving economics, not about the weights, and is exactly why the weights are archived rather than deleted. Band and generation are the run's own (min_pixels == max_pixels == 518400, PROMPT_VERSION 2), read from training_runs/20260906-025040/model.json rather than assumed."
+      "note": "A single-adapter Qwen3.8-27B. NOT the routed Abyss and not one of its arms — the base model is what a self-hoster loads, so it could never share that entry. Archived rather than served."
     },
     "CaptchaKraken/Abyss-grid": {
       "availability": "private",
@@ -224,7 +224,7 @@
         "min": 518400,
         "max": 518400
       },
-      "note": "PRIVATE on the Hub. The 'grid' expert arm of the routed CaptchaKraken/Abyss, from run 20260906-014318. Tier 2 gated_overall 0.7223 against a 0.6644 baseline, solved 0.6719, forgetting 0; Tier 3 9/10 pairs. Its run reads `blocked` for a TIER 1 failure that was this repo's own test harness (a module its sys.path insertion made unclassifiable, fixed in 79201f9aa) and not the model — the tier that judges the weights passed. It has its OWN entry rather than an alias onto the router because it now has its own weights at its own repo id, and `experts` names what goes ON THE WIRE — an alias would resolve a real, separately fetchable adapter to another model's entry. It still declares the router's prompt generation and pixel band, which is what check_prompt_parity.py requires of every expert and what makes the mixture coherent."
+      "note": "The 'grid' expert arm of the routed CaptchaKraken/Abyss. It has its own entry rather than an alias onto the router because it has its own weights at its own repo id, and `experts` names what goes on the wire. It declares the router's prompt generation and pixel band, which is what the parity gate requires of every arm and what makes the mixture coherent."
     },
     "CaptchaKraken/Abyss-text": {
       "availability": "private",
@@ -237,19 +237,45 @@
         "min": 518400,
         "max": 518400
       },
-      "note": "PRIVATE on the Hub. The 'text' expert arm of the routed CaptchaKraken/Abyss, from run 20260906-014253. Tier 2 gated_overall 0.8750 against a 0.8375 baseline, solved 0.8750, forgetting 0; Tier 3 5/6 pairs. `passed-partial` because the run is type-narrowed by construction — an arm trains on its own families — which disqualifies `ready-to-deploy` and says nothing bad about it. It has its OWN entry rather than an alias onto the router because it now has its own weights at its own repo id, and `experts` names what goes ON THE WIRE — an alias would resolve a real, separately fetchable adapter to another model's entry. It still declares the router's prompt generation and pixel band, which is what check_prompt_parity.py requires of every expert and what makes the mixture coherent."
+      "note": "The 'text' expert arm of the routed CaptchaKraken/Abyss. Own entry, own weights, and it declares the router's prompt generation and pixel band — see CaptchaKraken/Abyss-grid."
+    },
+    "CaptchaKraken/Abyss-video": {
+      "availability": "private",
+      "prompt_version": "2",
+      "kind": "lora",
+      "base_model": "Qwen/Qwen3.5-9B",
+      "lora_revision": "main",
+      "lora_name": "abyss-video",
+      "pixel_budget": {
+        "min": 518400,
+        "max": 518400
+      },
+      "note": "The 'video' expert arm of the routed CaptchaKraken/Abyss. Own entry, own weights, and it declares the router's prompt generation and pixel band — see CaptchaKraken/Abyss-grid."
+    },
+    "CaptchaKraken/Abyss-general": {
+      "availability": "private",
+      "prompt_version": "2",
+      "kind": "lora",
+      "base_model": "Qwen/Qwen3.5-9B",
+      "lora_revision": "main",
+      "lora_name": "abyss-general",
+      "pixel_budget": {
+        "min": 518400,
+        "max": 518400
+      },
+      "note": "The 'pixel' expert arm of the routed CaptchaKraken/Abyss and its generalist fallback: the router's own `lora_name` names this entry. Own entry, own weights, and it declares the router's prompt generation and pixel band — see CaptchaKraken/Abyss-grid."
     }
   },
   "served_aliases": {
-    "_comment": "A hosted endpoint serves a bare name, not a repo id. Map it back so prompt resolution works for callers who only ever see the alias. api.captchakraken.com now serves ONE model behind the permanent public name `captcha`, whatever the caller sends, so `captcha` means whichever adapter is current there — today CaptchaKraken-Lora-v1.2 — and NOT the v1.1 weights that share the name as a local vLLM alias. Self-hosters resolve by repo id via CAPTCHA_LORA_ADAPTER and are unaffected. Move this in the same change as the gateway's CKGATE_UPSTREAM_MODEL or the hosted model answers a prompt generation it was not trained on.",
+    "_comment": "A hosted endpoint serves a bare name, not a repo id. Map it back so prompt resolution works for callers who only ever see the alias. The hosted API serves one model behind the permanent public name `captcha`, whatever the caller sends, so `captcha` means whichever adapter is current there. Self-hosters resolve by repo id via CAPTCHA_LORA_ADAPTER and are unaffected. Move this in the same change as the gateway's upstream model, or the hosted model answers a prompt generation it was not trained on.",
     "captcha": "CaptchaKraken/CaptchaKraken-Lora-v1.2",
     "captcha-v12": "CaptchaKraken/CaptchaKraken-Lora-v1.2",
-    "_comment_abyss": "`abyss` is here for the OPPOSITE reason to the two above, and it is not a contradiction of `availability: licensed`. An alias says which model a served NAME means; availability says whether its weights can be downloaded. A licence holder does name this model — the hosted API accepts it from a licensed key and lists it to them on /v1/models so their SDK can — and without this line `canonical_model_id('abyss')` is None, so their prompts resolve through the `latest` fallback: the exact model/generation mispairing this file exists to prevent, silent, and only accidentally right while `latest` happens to share the generation. It does NOT make the weights fetchable; `captchakraken fetch` reads availability and refuses this name, which is also the clear error a self-hoster who typed it deserves.",
+    "_comment_abyss": "An alias says which model a served NAME means; `availability` says whether its weights can be downloaded. The two are independent. A licence holder names this model — the hosted API accepts it from a licensed key and lists it on /v1/models — and without this line prompt resolution falls through to the `latest` fallback, which is the model/generation mispairing this file exists to prevent. It does not make the weights fetchable.",
     "abyss": "CaptchaKraken/Abyss",
-    "_comment_abyss_experts": "The four expert arms. These are the names that go ON THE WIRE — a routed request for a grid names `abyss-grid`, never `abyss` — so without these lines `canonical_model_id` returns None for the string the client actually sent, and prompt resolution falls through to the `latest` fallback on every routed solve. That is the mispairing `abyss` itself is here to prevent, one level down and four times over. Aliasing four arms onto one entry rather than giving each its own is deliberate and CHECKED: they share a base, a prompt generation and a pixel band today, and check_prompt_parity.py requires every expert name to resolve to a registered entry declaring the SAME prompt_version and pixel_budget as the router — so the day an arm trains at another generation or another band, the gate refuses the alias and that arm has to graduate to an entry of its own.",
+    "_comment_abyss_experts": "The four expert arms. These are the names that go on the wire — a routed request for a grid names `abyss-grid`, never `abyss` — so without these lines prompt resolution falls through to `latest` on every routed solve. Aliasing the arms onto one entry is deliberate and checked: they share a base, a prompt generation and a pixel band, and the parity gate requires every expert name to resolve to a registered entry declaring the SAME prompt_version and pixel_budget as the router — so the day an arm trains at another generation or band, the gate refuses the alias and that arm has to graduate to an entry of its own.",
     "abyss-grid": "CaptchaKraken/Abyss-grid",
-    "abyss-general": "CaptchaKraken/Abyss",
-    "abyss-video": "CaptchaKraken/Abyss",
+    "abyss-general": "CaptchaKraken/Abyss-general",
+    "abyss-video": "CaptchaKraken/Abyss-video",
     "abyss-text": "CaptchaKraken/Abyss-text",
     "abyss-27b": "CaptchaKraken/Abyss-27B"
   }