feat(LTX25-GENERATED-KEYFRAMES): the second feature called "keyframe", refused by what is missing (#920)

.agents/issue-index.md

@@ -240,6 +240,7 @@ rather than merged. `scripts/check-agent-record.py` gates both.
 | [#907](https://github.com/mudler/vllm.cpp/issues/907) | `BACKEND-CUDA-COMP-CORE` | Five tests are red on dgx.casa (GB10, sm_121a) at `main`, proven PRE-EXISTING by a control build at `10b8bbdaa`: `test_capi` (SIGSEGV in an ABI v8 custom-logits-processor case, plausibly [#547](https://github.com/mudler/vllm.cpp/issues/547) or [#844](https://github.com/mudler/vllm.cpp/issues/844)), `test_cuda_ops` 439/440, `test_linear_method` 83/85, `test_ops_gdn` 4899/4900 ([#614](https://github.com/mudler/vllm.cpp/issues/614)), `test_qwen3_5_gdn_spec_routing` 119/123. Three of the five had no issue at all, which is why this exists | bug |
 | [#912](https://github.com/mudler/vllm.cpp/issues/912) | `ENG-EXPERT-STREAM` | Stream routed experts from NVMe so a model larger than device memory runs. Target `Qwen/Qwen3.8-2.4T-A95B`, REGISTERED against `Qwen3_5MoeForCausalLM` and blocked only on capacity: 370 GiB at UD-Q1_0 against 128 GB of unified memory on GB10. The only one of the three offload rows that helps on a unified-memory host, because `ENG-WEIGHT-OFFLOAD` and `ENG-HYBRID-PLACEMENT` both move bytes inside one physical pool. Cheaper than the spec assumed: on the GGUF path the mmap'd file already IS the bank and the per-expert slicer landed 2026-07-22 ([#824](https://github.com/mudler/vllm.cpp/issues/824)) | feature |
 | [#911](https://github.com/mudler/vllm.cpp/issues/911) | — | A `file:line` anchor into a file the row is ITSELF editing is stale by default, and spec BODIES are checked by nothing: `check-agent-record.py`'s `MATRIX_PATHS` (`:521`, `:529-530`) covers the five matrices, `feature-matrix.md` and `specs/model-family-inventory.md`, leaving 4772 line-carrying citations across 315 `.agents/specs/*.md` unexamined (positive control: 2314 line-less `.cpp` mentions match the same shape). `ltx25-prompt-adaln.md` shipped EIGHT stale repo-local anchors across two repair commits, moved by its own `020381676` and by `98f8e046d` (#658), then SEVEN more that were correct at `00613767d` and wrong at the merge of `origin/main`, because `0785cfc4d` (#882) added 70 lines to `ltx2_video.cpp` and 306 to `test_ltx2_video.cpp` ahead of every one. The obvious checker is a TAUTOLOGY — reading the span out of the file it validates reports 27 of 27 fresh on the same tree where reading the spans against their CLAIMS finds seven stale. Remedy is already in use and unwritten: `path:NN @ <sha>` for a historical claim, claim-sourced uniqueness re-derivation for a live one, re-run after the merge. Narrower than [#632](https://github.com/mudler/vllm.cpp/issues/632) on surface and sharper on mechanism. Listed under `## Owed` in [`ltx25-prompt-adaln.md`](specs/ltx25-prompt-adaln.md) | bug |
+| [#920](https://github.com/mudler/vllm.cpp/issues/920) | `LTX25-GENERATED-KEYFRAMES` | Generated keyframe slots — upstream `VideoGeneratedKeyframeSlots` (`conditioning/types/keyframe_slots.py:27-174` @ `fd4ded7f`), reached as `--num-generated-keyframes` (`ltx-pipelines/utils/args.py:833-844`) and documented at `ltx-pipelines/docs/conditioning.md:29-61` — have no REQUEST surface in this tree, so asking for them falls through the per-generation extras check and gets `unknown per-generation extra`, a message that says the family does not define the key and sends the reader looking for a typo. This is the ONLY upstream conditioning item that passes `marked=True` to `extend_keyframes_mask` (`:121` against `keyframe_cond.py:84-86`), so it is the only user-facing feature that puts #658's trained bias on a token other than the target's own first latent frame; `KeyframeInterpolationPipeline` does NOT (no `generated_keyframe` reference in its 362 lines, and absent from the feature's own applies-to list at `conditioning.md:47-51`). ONE blocker, the `GeneratedKeyframeLayout` READBACK: the slots are the OUTPUT, so they must be located by the layout the append recorded rather than assumed to trail, extracted before the extra tokens are trimmed, and each decoded as a STANDALONE one-frame clip (`ltx_core/types.py:269-272`). The token-append machinery is NOT a second blocker, and this row states that rather than restating the refusal it replaces: `c7cb59fbb` (#930) landed the seam and serves the LAST-frame supplied arm through it. Owed to THIS row rather than to #930 is a production caller for the `marked=true` branch of `Ltx2ExtendKeyframesMask` (`include/vllm/model_executor/models/ltx2_conditioning.h:175` @ `e5351776c`), whose only driver today is the unit case at `tests/vllm/models/test_ltx2_vae.cpp:2494` @ `e5351776c`. Spec [`ltx25-generated-keyframes.md`](specs/ltx25-generated-keyframes.md), campaign #644 | bug |
 | [#919](https://github.com/mudler/vllm.cpp/issues/919) | `LTX25-RESOLUTION-ENVELOPE` | `vllm_video_generate` integer-divides `width`/`height` into the latent grid (`ltx2_video.cpp:1456-1463 @ 5a0ffe9e3`) with no divisibility check, so a 100x100 request on the distilled two-stage recipe silently renders 96x96. The only geometry guard in the LTX path is a LOWER bound (`ltx2_video.cpp:1464-1471 @ 5a0ffe9e3`). Every repo-local anchor in this row is SHA-pinned because the fix edits the files it cites, and inserts lines above both spans (#911). Upstream hard-validates and raises at the top of a pipeline `__call__` — `assert_resolution` (`ltx-pipelines utils/helpers.py:540-551` @ `fd4ded7f2`), 64 for two-stage and 32 for one-stage, NINE invocations including `ti2vid_two_stages.py:184` and `ti2vid_two_stages_hq.py:199` — so mirroring means refusing, not flooring. Nine, not the 21 lines a grep for the name returns (9 invocations + 1 definition + 10 imports + 1 `__all__` string), and not every pipeline: 13 pipeline `__call__`s take a resolution and the three `*_mgpu.py` variants plus `hdr_ic_lora.py:352` skip the guard. `docs/USAGE.md:626-629 @ 5a0ffe9e3` already documents the rule as though it were enforced. Frames are the OPPOSITE answer: upstream floors an explicit `num_frames` exactly as we do (`ltx_core/types.py:113`) and validates it nowhere, so that half is a doc correction | bug |
 | [#921](https://github.com/mudler/vllm.cpp/issues/921) | — | The res_2s DENOISING LOOP (`ltx-pipelines utils/samplers.py:206-447` @ `fd4ded7f2`) is unported, so `TI2VidTwoStagesHQPipeline` cannot be served. What exists is one substep's SDE arithmetic: `Ltx2Res2sSdeCoeff`/`Ltx2Res2sStep` (`ltx2_pipeline.cpp:307-360 @ 5a0ffe9e3`, the two functions in full) mirror `Res2sDiffusionStep` (`diffusion_steps.py:118-190`) and are gated. Absent are the `phi`/`get_res2s_coefficients` exponential integrator (`utils/res2s.py:4-62`), the SECOND transformer evaluation per step at `sub_sigma = sqrt(sigma * sigma_next)` (`samplers.py:315` and `samplers.py:380-386`, spelt out because a bare `:NN` after a res2s.py citation reads as res2s.py) against our once-per-step loop (`ltx2_video.cpp:1735 @ 5a0ffe9e3`, with its single forward at `ltx2_video.cpp:1813-1817 @ 5a0ffe9e3`), the bong anchor refinement (`samplers.py:357-364`), and any `Ltx2StepperKind` enumerator to select it. The sampler IS the HQ variant, so this arm must refuse by name rather than substitute Euler and render something plausible that is quietly not HQ. Listed under `## Owed` in [`ltx25-resolution-envelope.md`](specs/ltx25-resolution-envelope.md) | feature |
 | [#922](https://github.com/mudler/vllm.cpp/issues/922) | `LTX25-A2V-AUDIO-INPUT` | LTX-2.5 audio-to-video (`A2VidPipelineTwoStage`, `a2vid_two_stage.py:53,143`) is absent: `vllm_video_params` carries no field or extra that accepts a driving waveform and `ltx2-gen` has no `--audio-path`, so nothing turns a file on disk into the audio latent the DiT's audio stream consumes. Distinct from reference-audio conditioning, which is correctly refused by name at `ltx2_video.cpp:1348-1355 @ 5a0ffe9e`; the two share one blocking dependency, the audio VAE ENCODER load path (`ltx2_loader.cpp:1295-1300` materializes `audio_vae.decoder.` only). The analysis half is already ported and unreached — `Ltx2AudioEncoderForward` (`ltx2_audio_vae.cpp:1114`), `Ltx2WaveformToLogMel` (`:1019`), `Ltx2SlaneyMelFilterbank` (`:970`) — and the engine applies ONE `phase.noise_scale` to both streams (`ltx2_video.cpp:1708-1712 @ 5a0ffe9e`) where upstream's `ModalitySpec` carries `noise_scale` and `frozen` per modality (`utils/types.py:99-112`). Spec [`ltx25-a2v-audio-input.md`](specs/ltx25-a2v-audio-input.md) | feature |