| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
fix(flux): address scheduler review follow-ups Correct Heun progress, scheduler reset, and explicit scheduler dispatch. Add regression coverage for LCM, previews, and CFG. | 5 天前 | |
Align DyPE with paper (#8980) | 5 个月前 | |
feat(working-memory): ship VAE working-memory coefficients as data (#9896) VAE estimators return a WorkingMemoryEstimate built from fitted and analytic terms; model_on_device accepts it as working_memory=. The AutoencoderKL coefficients and their measured points move to backend/working_memory/data, checked by a data test; reservations are unchanged. Co-authored-by: Lincoln Stein <lincoln.stein@gmail.com> | 6 小时前 | |
refactor: model manager v3 (#8607) * feat(mm): add UnknownModelConfig * refactor(ui): move model categorisation-ish logic to central location, simplify model manager models list * refactor(ui)refactor(ui): more cleanup of model categories * refactor(ui): remove unused excludeSubmodels I can't remember what this was for and don't see any reference to it. Maybe it's just remnants from a previous implementation? * feat(nodes): add unknown as model base * chore(ui): typegen * feat(ui): add unknown model base support in ui * feat(ui): allow changing model type in MM, fix up base and variant selects * feat(mm): omit model description instead of making it "base type filename model" * feat(app): add setting to allow unknown models * feat(ui): allow changing model format in MM * feat(app): add the installed model config to install complete events * chore(ui): typegen * feat(ui): toast warning when installed model is unidentified * docs: update config docstrings * chore(ui): typegen * tests(mm): fix test for MM, leave the UnknownModelConfig class in the list of configs * tidy(ui): prefer types from zod schemas for model attrs * chore(ui): lint * fix(ui): wrong translation string * feat(mm): normalized model storage Store models in a flat directory structure. Each model is in a dir named its unique key (a UUID). Inside that dir is either the model file or the model dir. * feat(mm): add migration to flat model storage * fix(mm): normalized multi-file/diffusers model installation no worky now worky * refactor: port MM probes to new api - Add concept of match certainty to new probe - Port CLIP Embed models to new API - Fiddle with stuff * feat(mm): port TIs to new API * tidy(mm): remove unused probes * feat(mm): port spandrel to new API * fix(mm): parsing for spandrel * fix(mm): loader for clip embed * fix(mm): tis use existing weight_files method * feat(mm): port vae to new API * fix(mm): vae class inheritance and config_path * tidy(mm): patcher types and import paths * feat(mm): better errors when invalid model config found in db * feat(mm): port t5 to new API * feat(mm): make config_path optional * refactor(mm): simplify model classification process Previously, we had a multi-phase strategy to identify models from their files on disk: 1. Run each model config classes' `matches()` method on the files. It checks if the model could possibly be an identified as the candidate model type. This was intended to be a quick check. Break on the first match. 2. If we have a match, run the config class's `parse()` method. It derive some additional model config attrs from the model files. This was intended to encapsulate heavier operations that may require loading the model into memory. 3. Derive the common model config attrs, like name, description, calculate the hash, etc. Some of these are also heavier operations. This strategy has some issues: - It is not clear how the pieces fit together. There is some back-and-forth between different methods and the config base class. It is hard to trace the flow of logic until you fully wrap your head around the system and therefore difficult to add a model architecture to the probe. - The assumption that we could do quick, lightweight checks before heavier checks is incorrect. We often _must_ load the model state dict in the `matches()` method. So there is no practical perf benefit to splitting up the responsibility of `matches()` and `parse()`. - Sometimes we need to do the same checks in `matches()` and `parse()`. In these cases, splitting the logic is has a negative perf impact because we are doing the same work twice. - As we introduce the concept of an "unknown" model config (i.e. a model that we cannot identify, but still record in the db; see #8582), we will _always_ run _all_ the checks for every model. Therefore we need not try to defer heavier checks or resource-intensive ops like hashing. We are going to do them anyways. - There are situations where a model may match multiple configs. One known case are SD pipeline models with merged LoRAs. In the old probe API, we relied on the implicit order of checks to know that if a model matched for pipeline _and_ LoRA, we prefer the pipeline match. But, in the new API, we do not have this implicit ordering of checks. To resolve this in a resilient way, we need to get all matches up front, then use tie-breaker logic to figure out which should win (or add "differential diagnosis" logic to the matchers). - Field overrides weren't handled well by this strategy. They were only applied at the very end, if a model matched successfully. This means we cannot tell the system "Hey, this model is type X with base Y. Trust me bro.". We cannot override the match logic. As we move towards letting users correct mis-identified models (see #8582), this is a requirement. We can simplify the process significantly and better support "unknown" models. Firstly, model config classes now have a single `from_model_on_disk()` method that attempts to construct an instance of the class from the model files. This replaces the `matches()` and `parse()` methods. If we fail to create the config instance, a special exception is raised that indicates why we think the files cannot be identified as the given model config class. Next, the flow for model identification is a bit simpler: - Derive all the common fields up-front (name, desc, hash, etc). - Merge in overrides. - Call `from_model_on_disk()` for every config class, passing in the fields. Overrides are handled in this method. - Record the results for each config class and choose the best one. The identification logic is a bit more verbose, with the special exceptions and handling of overrides, but it is very clear what is happening. The one downside I can think of for this strategy is we do need to check every model type, instead of stopping at the first match. It's a bit less efficient. In practice, however, this isn't a hot code path, and the improved clarity is worth far more than perf optimizations that the end user will likely never notice. * refactor(mm): remove unused methods in config.py * refactor(mm): add model config parsing utils * fix(mm): abstractmethod bork * tidy(mm): clarify that model id utils are private * fix(mm): fall back to UnknownModelConfig correctly * feat(mm): port CLIPVisionDiffusersConfig to new api * feat(mm): port SigLIPDiffusersConfig to new api * feat(mm): make match helpers more succint * feat(mm): port flux redux to new api * feat(mm): port ip adapter to new api * tidy(mm): skip optimistic override handling for now * refactor(mm): continue iterating on config * feat(mm): port flux "control lora" and t2i adapter to new api * tidy(ui): use Extract to get model config types * fix(mm): t2i base determination * feat(mm): port cnet to new api * refactor(mm): add config validation utils, make it all consistent and clean * feat(mm): wip port of main models to new api * feat(mm): wip port of main models to new api * feat(mm): wip port of main models to new api * docs(mm): add todos * tidy(mm): removed unused model merge class * feat(mm): wip port main models to new api * tidy(mm): clean up model heuristic utils * tidy(mm): clean up ModelOnDisk caching * tidy(mm): flux lora format util * refactor(mm): make config classes narrow Simpler logic to identify, less complexity to add new model, fewer useless attrs that do not relate to the model arch, etc * refactor(mm): diffusers loras w * feat(mm): consistent naming for all model config classes * fix(mm): tag generation & scattered probe fixes * tidy(mm): consistent class names * refactor(mm): split configs into separate files * docs(mm): add comments for identification utils * chore(ui): typegen * refactor(mm): remove legacy probe, new configs dir structure, update imports * fix(mm): inverted condition * docs(mm): update docsstrings in factory.py * docs(mm): document flux variant attr * feat(mm): add helper method for legacy configs * feat(mm): satisfy type checker in flux denoise * docs(mm): remove extraneous comment * fix(mm): ensure unknown model configs get unknown attrs * fix(mm): t5 identification * fix(mm): sdxl ip adapter identification * feat(mm): more flexible config matching utils * fix(mm): clip vision identification * feat(mm): add sanity checks before probing paths * docs(mm): add reminder for self for field migrations * feat(mm): clearer naming for main config class hierarchy * feat(mm): fix clip vision starter model bases, add ref to actual models * feat(mm): add model config schema migration logic * fix(mm): duplicate import * refactor(mm): split big migration into 3 Split the big migration that did all of these things into 3: - Migration 22: Remove unique contraint on base/name/type in models table - Migration 23: Migrate configs to v6.8.0 schemas - Migration 24: Normalize file storage * fix(mm): pop base/type/format when creating unknown model config * fix(db): migration 22 insert only real cols * fix(db): migration 23 fall back to unknown model when config change fails * feat(db): run migrations 23 and 24 * fix(mm): false negative on flux lora * fix(mm): vae checkpoint probe checking for dir instead of file * fix(mm): ModelOnDisk skips dirs when looking for weights Previously a path w/ any of the known weights suffixes would be seen as a weights file, even if it was a directory. We now check to ensure the candidate path is actually a file before adding it to the list of weights. * feat(mm): add method to get main model defaults from a base * feat(mm): do not log when multiple non-unknown model matches * refactor(mm): continued iteration on model identifcation * tests(mm): refactor model identification tests Overhaul of model identification (probing) tests. Previously we didn't test the correctness of probing except in a few narrow cases - now we do. See tests/model_identification/README.md for a detailed overview of the new test setup. It includes instructions for adding a new test case. In brief: - Download the model you want to add as a test case - Run a script against it to generate the test model files - Fill in the expected model type/format/base/etc in the generated test metadata JSON file Included test cases: - All starter models - A handful of other models that I had installed - Models present in the previous test cases as smoke tests, now also tested for correctness * fix(mm): omit type/format/base when creating unknown config instance * feat(mm): use ValueError for model id sanity checks * feat(mm): add flag for updating models to allow class changes * tests(mm): fix remaining MM tests * feat: allow users to edit models freely * feat(ui): add warning for model settings edit * tests(mm): flux state dict tests * tidy: remove unused file * fix(mm): lora state dict loading in model id * feat(ui): use translation string for model edit warning * docs(db): update version numbers in migration comments * chore: bump version to v6.9.0a1 * docs: update model id readme * tests(mm): attempt to fix windows model id tests * fix(mm): issue with deleting single file models * feat(mm): just delete the dir w/ rmtree when deleting model * tests(mm): windows CI issue * fix(ui): typegen schema sync * fix(mm): fixes for migration 23 - Handle CLIP Embed and Main SD models missing variant field - Handle errors when calling the discriminator function, previously only handled ValidationError but it could be a ValueError or something else - Better logging for config migration * chore: bump version to v6.9.0a2 * chore: bump version to v6.9.0a3 | 11 个月前 | |
feat(flux): load the FLUX.1 VAE as a diffusers AutoencoderKL, and tile its encode An upscale graph hands i2l the frame after Spandrel has enlarged it, and FLUX.1 could not encode that frame: its VAE was InvokeAI's port of the BFL reference, which implemented a tiled decode and never a tiled encode. Measured on a 4090, bf16: a 3072px encode reserves 18.84 GiB untiled against 0.81 GiB tiled, and the tiled pass is the faster of the two because the untiled one is quadratic in the mid-block attention. On a 16 GiB card it does not finish at all. The diffusers class is the same network, not a similar one: fp32 maxdiff 1.6e-06 on a latent and 1.5e-05 on an image, at corr 1.000000000. It brings tiling in both directions and removes 513 lines of vendored autoencoder. Only the BFL key layout is accepted. A standalone diffusers-layout file carries no config, and SD 3.5's and CogView 4's autoencoders are the same 244 tensors with the same shapes, differing only in the normalisation -- while identification labels every 16-channel VAE as flux. Loading one would stamp FLUX's constants onto foreign weights: it loads, it generates, and the image is wrong with nothing to say so. It is refused with that reason. Also here: tiled/tile_size on flux_vae_encode, flux_vae_decode, z_image_i2l and anima_i2l (replacing a hard-coded disable_tiling()); an encode-shaped residual in the working-memory estimator, since a tiled encode does not bound the input frame or the moments it assembles; and an OOM-to-tiled fallback on the FLUX and Kontext encodes, the latter because nothing on that path resizes the user's reference image. Every FLUX.1 and Z-Image image shifts by bf16 rounding: rmse 0.00167 on real content, ten times smaller than the VAE's own round-trip loss. Seamless now applies to a standalone FLUX VAE, where it was silently skipped. | 21 天前 | |
refactor: model manager v3 (#8607) * feat(mm): add UnknownModelConfig * refactor(ui): move model categorisation-ish logic to central location, simplify model manager models list * refactor(ui)refactor(ui): more cleanup of model categories * refactor(ui): remove unused excludeSubmodels I can't remember what this was for and don't see any reference to it. Maybe it's just remnants from a previous implementation? * feat(nodes): add unknown as model base * chore(ui): typegen * feat(ui): add unknown model base support in ui * feat(ui): allow changing model type in MM, fix up base and variant selects * feat(mm): omit model description instead of making it "base type filename model" * feat(app): add setting to allow unknown models * feat(ui): allow changing model format in MM * feat(app): add the installed model config to install complete events * chore(ui): typegen * feat(ui): toast warning when installed model is unidentified * docs: update config docstrings * chore(ui): typegen * tests(mm): fix test for MM, leave the UnknownModelConfig class in the list of configs * tidy(ui): prefer types from zod schemas for model attrs * chore(ui): lint * fix(ui): wrong translation string * feat(mm): normalized model storage Store models in a flat directory structure. Each model is in a dir named its unique key (a UUID). Inside that dir is either the model file or the model dir. * feat(mm): add migration to flat model storage * fix(mm): normalized multi-file/diffusers model installation no worky now worky * refactor: port MM probes to new api - Add concept of match certainty to new probe - Port CLIP Embed models to new API - Fiddle with stuff * feat(mm): port TIs to new API * tidy(mm): remove unused probes * feat(mm): port spandrel to new API * fix(mm): parsing for spandrel * fix(mm): loader for clip embed * fix(mm): tis use existing weight_files method * feat(mm): port vae to new API * fix(mm): vae class inheritance and config_path * tidy(mm): patcher types and import paths * feat(mm): better errors when invalid model config found in db * feat(mm): port t5 to new API * feat(mm): make config_path optional * refactor(mm): simplify model classification process Previously, we had a multi-phase strategy to identify models from their files on disk: 1. Run each model config classes' `matches()` method on the files. It checks if the model could possibly be an identified as the candidate model type. This was intended to be a quick check. Break on the first match. 2. If we have a match, run the config class's `parse()` method. It derive some additional model config attrs from the model files. This was intended to encapsulate heavier operations that may require loading the model into memory. 3. Derive the common model config attrs, like name, description, calculate the hash, etc. Some of these are also heavier operations. This strategy has some issues: - It is not clear how the pieces fit together. There is some back-and-forth between different methods and the config base class. It is hard to trace the flow of logic until you fully wrap your head around the system and therefore difficult to add a model architecture to the probe. - The assumption that we could do quick, lightweight checks before heavier checks is incorrect. We often _must_ load the model state dict in the `matches()` method. So there is no practical perf benefit to splitting up the responsibility of `matches()` and `parse()`. - Sometimes we need to do the same checks in `matches()` and `parse()`. In these cases, splitting the logic is has a negative perf impact because we are doing the same work twice. - As we introduce the concept of an "unknown" model config (i.e. a model that we cannot identify, but still record in the db; see #8582), we will _always_ run _all_ the checks for every model. Therefore we need not try to defer heavier checks or resource-intensive ops like hashing. We are going to do them anyways. - There are situations where a model may match multiple configs. One known case are SD pipeline models with merged LoRAs. In the old probe API, we relied on the implicit order of checks to know that if a model matched for pipeline _and_ LoRA, we prefer the pipeline match. But, in the new API, we do not have this implicit ordering of checks. To resolve this in a resilient way, we need to get all matches up front, then use tie-breaker logic to figure out which should win (or add "differential diagnosis" logic to the matchers). - Field overrides weren't handled well by this strategy. They were only applied at the very end, if a model matched successfully. This means we cannot tell the system "Hey, this model is type X with base Y. Trust me bro.". We cannot override the match logic. As we move towards letting users correct mis-identified models (see #8582), this is a requirement. We can simplify the process significantly and better support "unknown" models. Firstly, model config classes now have a single `from_model_on_disk()` method that attempts to construct an instance of the class from the model files. This replaces the `matches()` and `parse()` methods. If we fail to create the config instance, a special exception is raised that indicates why we think the files cannot be identified as the given model config class. Next, the flow for model identification is a bit simpler: - Derive all the common fields up-front (name, desc, hash, etc). - Merge in overrides. - Call `from_model_on_disk()` for every config class, passing in the fields. Overrides are handled in this method. - Record the results for each config class and choose the best one. The identification logic is a bit more verbose, with the special exceptions and handling of overrides, but it is very clear what is happening. The one downside I can think of for this strategy is we do need to check every model type, instead of stopping at the first match. It's a bit less efficient. In practice, however, this isn't a hot code path, and the improved clarity is worth far more than perf optimizations that the end user will likely never notice. * refactor(mm): remove unused methods in config.py * refactor(mm): add model config parsing utils * fix(mm): abstractmethod bork * tidy(mm): clarify that model id utils are private * fix(mm): fall back to UnknownModelConfig correctly * feat(mm): port CLIPVisionDiffusersConfig to new api * feat(mm): port SigLIPDiffusersConfig to new api * feat(mm): make match helpers more succint * feat(mm): port flux redux to new api * feat(mm): port ip adapter to new api * tidy(mm): skip optimistic override handling for now * refactor(mm): continue iterating on config * feat(mm): port flux "control lora" and t2i adapter to new api * tidy(ui): use Extract to get model config types * fix(mm): t2i base determination * feat(mm): port cnet to new api * refactor(mm): add config validation utils, make it all consistent and clean * feat(mm): wip port of main models to new api * feat(mm): wip port of main models to new api * feat(mm): wip port of main models to new api * docs(mm): add todos * tidy(mm): removed unused model merge class * feat(mm): wip port main models to new api * tidy(mm): clean up model heuristic utils * tidy(mm): clean up ModelOnDisk caching * tidy(mm): flux lora format util * refactor(mm): make config classes narrow Simpler logic to identify, less complexity to add new model, fewer useless attrs that do not relate to the model arch, etc * refactor(mm): diffusers loras w * feat(mm): consistent naming for all model config classes * fix(mm): tag generation & scattered probe fixes * tidy(mm): consistent class names * refactor(mm): split configs into separate files * docs(mm): add comments for identification utils * chore(ui): typegen * refactor(mm): remove legacy probe, new configs dir structure, update imports * fix(mm): inverted condition * docs(mm): update docsstrings in factory.py * docs(mm): document flux variant attr * feat(mm): add helper method for legacy configs * feat(mm): satisfy type checker in flux denoise * docs(mm): remove extraneous comment * fix(mm): ensure unknown model configs get unknown attrs * fix(mm): t5 identification * fix(mm): sdxl ip adapter identification * feat(mm): more flexible config matching utils * fix(mm): clip vision identification * feat(mm): add sanity checks before probing paths * docs(mm): add reminder for self for field migrations * feat(mm): clearer naming for main config class hierarchy * feat(mm): fix clip vision starter model bases, add ref to actual models * feat(mm): add model config schema migration logic * fix(mm): duplicate import * refactor(mm): split big migration into 3 Split the big migration that did all of these things into 3: - Migration 22: Remove unique contraint on base/name/type in models table - Migration 23: Migrate configs to v6.8.0 schemas - Migration 24: Normalize file storage * fix(mm): pop base/type/format when creating unknown model config * fix(db): migration 22 insert only real cols * fix(db): migration 23 fall back to unknown model when config change fails * feat(db): run migrations 23 and 24 * fix(mm): false negative on flux lora * fix(mm): vae checkpoint probe checking for dir instead of file * fix(mm): ModelOnDisk skips dirs when looking for weights Previously a path w/ any of the known weights suffixes would be seen as a weights file, even if it was a directory. We now check to ensure the candidate path is actually a file before adding it to the list of weights. * feat(mm): add method to get main model defaults from a base * feat(mm): do not log when multiple non-unknown model matches * refactor(mm): continued iteration on model identifcation * tests(mm): refactor model identification tests Overhaul of model identification (probing) tests. Previously we didn't test the correctness of probing except in a few narrow cases - now we do. See tests/model_identification/README.md for a detailed overview of the new test setup. It includes instructions for adding a new test case. In brief: - Download the model you want to add as a test case - Run a script against it to generate the test model files - Fill in the expected model type/format/base/etc in the generated test metadata JSON file Included test cases: - All starter models - A handful of other models that I had installed - Models present in the previous test cases as smoke tests, now also tested for correctness * fix(mm): omit type/format/base when creating unknown config instance * feat(mm): use ValueError for model id sanity checks * feat(mm): add flag for updating models to allow class changes * tests(mm): fix remaining MM tests * feat: allow users to edit models freely * feat(ui): add warning for model settings edit * tests(mm): flux state dict tests * tidy: remove unused file * fix(mm): lora state dict loading in model id * feat(ui): use translation string for model edit warning * docs(db): update version numbers in migration comments * chore: bump version to v6.9.0a1 * docs: update model id readme * tests(mm): attempt to fix windows model id tests * fix(mm): issue with deleting single file models * feat(mm): just delete the dir w/ rmtree when deleting model * tests(mm): windows CI issue * fix(ui): typegen schema sync * fix(mm): fixes for migration 23 - Handle CLIP Embed and Main SD models missing variant field - Handle errors when calling the discriminator function, previously only handled ValidationError but it could be a ValueError or something else - Better logging for config migration * chore: bump version to v6.9.0a2 * chore: bump version to v6.9.0a3 | 11 个月前 | |
Distinguish between restricted and unrestricted attn masks in FLUX regional prompting. | 1 年前 | |
fix(flux): cover the LCM step generator at the invocation level Address review on #9837: - Add invocation-level tests that drive the real _run_diffusion for FLUX.1, FLUX.2 and Z-Image and assert the step generator is a CPU generator seeded from seed ^ 0xFFFFFFFF. Dropping generator= from either FLUX node, or the Z-Image step_generator, now fails them. - Move the Z-Image step generator and its comment above the scheduler stepping comments, so the tqdm comment sits next to the tqdm call. - Reword the backend comments: Euler and Heun do not use the step noise here (s_churn is 0), rather than not consulting the generator. Signed-off-by: GokayAI <60583610+gokay-ai@users.noreply.github.com> | 2 天前 | |
refactor: model manager v3 (#8607) * feat(mm): add UnknownModelConfig * refactor(ui): move model categorisation-ish logic to central location, simplify model manager models list * refactor(ui)refactor(ui): more cleanup of model categories * refactor(ui): remove unused excludeSubmodels I can't remember what this was for and don't see any reference to it. Maybe it's just remnants from a previous implementation? * feat(nodes): add unknown as model base * chore(ui): typegen * feat(ui): add unknown model base support in ui * feat(ui): allow changing model type in MM, fix up base and variant selects * feat(mm): omit model description instead of making it "base type filename model" * feat(app): add setting to allow unknown models * feat(ui): allow changing model format in MM * feat(app): add the installed model config to install complete events * chore(ui): typegen * feat(ui): toast warning when installed model is unidentified * docs: update config docstrings * chore(ui): typegen * tests(mm): fix test for MM, leave the UnknownModelConfig class in the list of configs * tidy(ui): prefer types from zod schemas for model attrs * chore(ui): lint * fix(ui): wrong translation string * feat(mm): normalized model storage Store models in a flat directory structure. Each model is in a dir named its unique key (a UUID). Inside that dir is either the model file or the model dir. * feat(mm): add migration to flat model storage * fix(mm): normalized multi-file/diffusers model installation no worky now worky * refactor: port MM probes to new api - Add concept of match certainty to new probe - Port CLIP Embed models to new API - Fiddle with stuff * feat(mm): port TIs to new API * tidy(mm): remove unused probes * feat(mm): port spandrel to new API * fix(mm): parsing for spandrel * fix(mm): loader for clip embed * fix(mm): tis use existing weight_files method * feat(mm): port vae to new API * fix(mm): vae class inheritance and config_path * tidy(mm): patcher types and import paths * feat(mm): better errors when invalid model config found in db * feat(mm): port t5 to new API * feat(mm): make config_path optional * refactor(mm): simplify model classification process Previously, we had a multi-phase strategy to identify models from their files on disk: 1. Run each model config classes' `matches()` method on the files. It checks if the model could possibly be an identified as the candidate model type. This was intended to be a quick check. Break on the first match. 2. If we have a match, run the config class's `parse()` method. It derive some additional model config attrs from the model files. This was intended to encapsulate heavier operations that may require loading the model into memory. 3. Derive the common model config attrs, like name, description, calculate the hash, etc. Some of these are also heavier operations. This strategy has some issues: - It is not clear how the pieces fit together. There is some back-and-forth between different methods and the config base class. It is hard to trace the flow of logic until you fully wrap your head around the system and therefore difficult to add a model architecture to the probe. - The assumption that we could do quick, lightweight checks before heavier checks is incorrect. We often _must_ load the model state dict in the `matches()` method. So there is no practical perf benefit to splitting up the responsibility of `matches()` and `parse()`. - Sometimes we need to do the same checks in `matches()` and `parse()`. In these cases, splitting the logic is has a negative perf impact because we are doing the same work twice. - As we introduce the concept of an "unknown" model config (i.e. a model that we cannot identify, but still record in the db; see #8582), we will _always_ run _all_ the checks for every model. Therefore we need not try to defer heavier checks or resource-intensive ops like hashing. We are going to do them anyways. - There are situations where a model may match multiple configs. One known case are SD pipeline models with merged LoRAs. In the old probe API, we relied on the implicit order of checks to know that if a model matched for pipeline _and_ LoRA, we prefer the pipeline match. But, in the new API, we do not have this implicit ordering of checks. To resolve this in a resilient way, we need to get all matches up front, then use tie-breaker logic to figure out which should win (or add "differential diagnosis" logic to the matchers). - Field overrides weren't handled well by this strategy. They were only applied at the very end, if a model matched successfully. This means we cannot tell the system "Hey, this model is type X with base Y. Trust me bro.". We cannot override the match logic. As we move towards letting users correct mis-identified models (see #8582), this is a requirement. We can simplify the process significantly and better support "unknown" models. Firstly, model config classes now have a single `from_model_on_disk()` method that attempts to construct an instance of the class from the model files. This replaces the `matches()` and `parse()` methods. If we fail to create the config instance, a special exception is raised that indicates why we think the files cannot be identified as the given model config class. Next, the flow for model identification is a bit simpler: - Derive all the common fields up-front (name, desc, hash, etc). - Merge in overrides. - Call `from_model_on_disk()` for every config class, passing in the fields. Overrides are handled in this method. - Record the results for each config class and choose the best one. The identification logic is a bit more verbose, with the special exceptions and handling of overrides, but it is very clear what is happening. The one downside I can think of for this strategy is we do need to check every model type, instead of stopping at the first match. It's a bit less efficient. In practice, however, this isn't a hot code path, and the improved clarity is worth far more than perf optimizations that the end user will likely never notice. * refactor(mm): remove unused methods in config.py * refactor(mm): add model config parsing utils * fix(mm): abstractmethod bork * tidy(mm): clarify that model id utils are private * fix(mm): fall back to UnknownModelConfig correctly * feat(mm): port CLIPVisionDiffusersConfig to new api * feat(mm): port SigLIPDiffusersConfig to new api * feat(mm): make match helpers more succint * feat(mm): port flux redux to new api * feat(mm): port ip adapter to new api * tidy(mm): skip optimistic override handling for now * refactor(mm): continue iterating on config * feat(mm): port flux "control lora" and t2i adapter to new api * tidy(ui): use Extract to get model config types * fix(mm): t2i base determination * feat(mm): port cnet to new api * refactor(mm): add config validation utils, make it all consistent and clean * feat(mm): wip port of main models to new api * feat(mm): wip port of main models to new api * feat(mm): wip port of main models to new api * docs(mm): add todos * tidy(mm): removed unused model merge class * feat(mm): wip port main models to new api * tidy(mm): clean up model heuristic utils * tidy(mm): clean up ModelOnDisk caching * tidy(mm): flux lora format util * refactor(mm): make config classes narrow Simpler logic to identify, less complexity to add new model, fewer useless attrs that do not relate to the model arch, etc * refactor(mm): diffusers loras w * feat(mm): consistent naming for all model config classes * fix(mm): tag generation & scattered probe fixes * tidy(mm): consistent class names * refactor(mm): split configs into separate files * docs(mm): add comments for identification utils * chore(ui): typegen * refactor(mm): remove legacy probe, new configs dir structure, update imports * fix(mm): inverted condition * docs(mm): update docsstrings in factory.py * docs(mm): document flux variant attr * feat(mm): add helper method for legacy configs * feat(mm): satisfy type checker in flux denoise * docs(mm): remove extraneous comment * fix(mm): ensure unknown model configs get unknown attrs * fix(mm): t5 identification * fix(mm): sdxl ip adapter identification * feat(mm): more flexible config matching utils * fix(mm): clip vision identification * feat(mm): add sanity checks before probing paths * docs(mm): add reminder for self for field migrations * feat(mm): clearer naming for main config class hierarchy * feat(mm): fix clip vision starter model bases, add ref to actual models * feat(mm): add model config schema migration logic * fix(mm): duplicate import * refactor(mm): split big migration into 3 Split the big migration that did all of these things into 3: - Migration 22: Remove unique contraint on base/name/type in models table - Migration 23: Migrate configs to v6.8.0 schemas - Migration 24: Normalize file storage * fix(mm): pop base/type/format when creating unknown model config * fix(db): migration 22 insert only real cols * fix(db): migration 23 fall back to unknown model when config change fails * feat(db): run migrations 23 and 24 * fix(mm): false negative on flux lora * fix(mm): vae checkpoint probe checking for dir instead of file * fix(mm): ModelOnDisk skips dirs when looking for weights Previously a path w/ any of the known weights suffixes would be seen as a weights file, even if it was a directory. We now check to ensure the candidate path is actually a file before adding it to the list of weights. * feat(mm): add method to get main model defaults from a base * feat(mm): do not log when multiple non-unknown model matches * refactor(mm): continued iteration on model identifcation * tests(mm): refactor model identification tests Overhaul of model identification (probing) tests. Previously we didn't test the correctness of probing except in a few narrow cases - now we do. See tests/model_identification/README.md for a detailed overview of the new test setup. It includes instructions for adding a new test case. In brief: - Download the model you want to add as a test case - Run a script against it to generate the test model files - Fill in the expected model type/format/base/etc in the generated test metadata JSON file Included test cases: - All starter models - A handful of other models that I had installed - Models present in the previous test cases as smoke tests, now also tested for correctness * fix(mm): omit type/format/base when creating unknown config instance * feat(mm): use ValueError for model id sanity checks * feat(mm): add flag for updating models to allow class changes * tests(mm): fix remaining MM tests * feat: allow users to edit models freely * feat(ui): add warning for model settings edit * tests(mm): flux state dict tests * tidy: remove unused file * fix(mm): lora state dict loading in model id * feat(ui): use translation string for model edit warning * docs(db): update version numbers in migration comments * chore: bump version to v6.9.0a1 * docs: update model id readme * tests(mm): attempt to fix windows model id tests * fix(mm): issue with deleting single file models * feat(mm): just delete the dir w/ rmtree when deleting model * tests(mm): windows CI issue * fix(ui): typegen schema sync * fix(mm): fixes for migration 23 - Handle CLIP Embed and Main SD models missing variant field - Handle errors when calling the discriminator function, previously only handled ValidationError but it could be a ValueError or something else - Better logging for config migration * chore: bump version to v6.9.0a2 * chore: bump version to v6.9.0a3 | 11 个月前 | |
Revert performance improvement, because it caused flux inference to fail on Mac: https://github.com/invoke-ai/InvokeAI/issues/7422 | 1 年前 | |
Rename Structural Lora to Control Lora | 1 年前 | |
Merge remote-tracking branch 'origin/main' into chore/torch-2.13 Carries the branch's regenerated API contracts to their moved location under frontend/api. | 17 天前 | |
fix(flux): address scheduler review follow-ups Correct Heun progress, scheduler reset, and explicit scheduler dispatch. Add regression coverage for LCM, previews, and CFG. | 5 天前 | |
Add support for FLUX Redux masks. | 1 年前 | |
fix(model-manager): identify standalone 16-channel VAEs as FLUX.1 or SD3 Folders are identified by their scaling/shift constants and single files by an explicit base, then the backbone their name or install source states; adds FLUX.1-diffusers and SD3 VAE configs, and an override now decides within a latent family instead of being overruled. VAELoader loads SD3 single files in either layout, FLUX.1 folders outside float16, and flat VAE folders requested as SubModelType.VAE. Reidentify passes the stored install source, so name evidence survives a re-probe. | 20 天前 |
| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
| 5 天前 | ||
| 5 个月前 | ||
| 6 小时前 | ||
| 11 个月前 | ||
| 21 天前 | ||
| 11 个月前 | ||
| 1 年前 | ||
| 2 天前 | ||
| 11 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 17 天前 | ||
| 5 天前 | ||
| 1 年前 | ||
| 20 天前 |