🤖 Auto-update Model Catalog - 24-11-2025 13:06

[jan-service-account]

🤖 Automated Model Catalog Update

Date: 24-11-2025 13:06

Summary

• New models added: 710
• Total models in catalog: 699

---

Important

Automated update adds 710 new models to the catalog, updates existing entries with missing keys, and finalizes the catalog size at 700 models.

• Behavior:
  • Adds 710 new models to the catalog, increasing the total to 700 models.
  • Updates existing models with missing keys, ensuring complete metadata.
  • Processes models from the Hugging Face API and handles existing models not found in the API.
• Catalog Management:
  • Final catalog size is 700 models.
  • Ensures all models have complete metadata by fetching missing keys where possible.
  • Removes models with multi-part GGUF files from the catalog.

^{This description was created by for bfa97dc. You can customize this summary. It will automatically update as commits are pushed.}

[ellipsis-dev]

Important

Looks good to me! 👍

Reviewed everything up to bfa97dc in 14 minutes and 17 seconds. Click for details.

• Reviewed 7861 lines of code in 1 files
• Skipped 1 files when reviewing.
• Skipped posting 14 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. script_output.log:50

• Draft comment:
  Repeated log lines like '-> Missing keys: ['isGated']' are added for many models. Consider refactoring or grouping these checks to reduce verbosity.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% This is a log file diff, not source code. The comment is suggesting refactoring the logging output to be less verbose. However, this is a script output log file - it's the OUTPUT of a script, not the script itself. The comment seems to be asking to change how the script logs information, which would require changes to the actual script code (not shown in this diff). The diff only shows the log output changes. Additionally, the evidence provided doesn't even show the specific pattern mentioned in the comment. This comment is not actionable based on this diff alone since we can't modify the logging behavior by editing a log file. Could the comment be valid if we interpret it as feedback about the script that generated this log? Perhaps the reviewer is looking at the log output and suggesting improvements to the underlying script. However, the rules state we should only comment on changes in the diff itself, and this is just a log file output. The rules explicitly state "Do NOT comment unless there is clearly a code change required" and "Ignore cross-file issues. Only think about the file you are reviewing." This is a log file, not source code. You cannot refactor a log file - you can only refactor the code that generates it. The comment is not actionable within the context of this diff. Additionally, the evidence provided doesn't even demonstrate the issue mentioned in the comment. This comment should be deleted. It's commenting on a log file output rather than source code, making it not actionable. The suggestion to "refactor or group these checks" cannot be implemented by modifying a log file. The comment would need to be on the actual script file that generates this log.

2. script_output.log:200

• Draft comment:
  Log formatting is inconsistent. Some lines say 'Updated existing entry' while others add 'with missing keys'. Standardize the message format for clarity.
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

3. script_output.log:400

• Draft comment:
  Several models show 404 error messages (e.g., 'Model not accessible via API'). Ensure error handling and messaging are consistent across entries.
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

4. script_output.log:600

• Draft comment:
  The log output for 'Processing existing model not in API:' is very detailed and repetitive. Consider grouping similar log messages to reduce noise.
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

5. script_output.log:108

• Draft comment:
  Typo alert: The repository name "facebook/sam-3d-body-vith" might be a misspelling. Please verify if it should be "...with" instead of "vith".
• Reason this comment was not posted:
  Comment was on unchanged code.

6. script_output.log:228

• Draft comment:
  Typo alert: The repository name "Kwai-Kolors/CoTyle" might contain a typo. Confirm whether "CoTyle" is correct or if it should be "CoStyle".
• Reason this comment was not posted:
  Comment was on unchanged code.

7. script_output.log:295

• Draft comment:
  Typo alert: The word "abliterated" in "OpenAi-GPT-oss-20b-abliterated-uncensored-NEO-Imatrix-gguf" may be misspelled. Please check if it should be "obliterated".
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 0% vs. threshold = 50% "Abliterated" is not a typo - it's intentional technical jargon in the AI/ML community. The term refers to a specific technique for removing alignment/safety features from models. This is widely used terminology in model names on Hugging Face, particularly for uncensored or modified models. The comment is incorrect and would mislead the PR author into thinking there's an error when there isn't one. This comment should be deleted as it's based on a misunderstanding of domain-specific terminology. Could there be a scenario where this is actually a typo in this specific instance, even if "abliterated" is valid terminology elsewhere? Perhaps the model author made a mistake? However, given that this term appears consistently across many different model repositories and authors in the log, and it's a well-established term in the community, this is extremely unlikely. While it's theoretically possible this could be a typo in one specific case, the overwhelming evidence shows "abliterated" is intentional technical terminology used consistently across the AI/ML community. The comment would cause confusion by suggesting a correction that would actually be wrong. This is clearly not a typo. Delete this comment. "Abliterated" is correct technical terminology in the AI/ML field, not a typo. The comment is incorrect and would mislead the PR author.

8. script_output.log:368

• Draft comment:
  Typo suggestion: In 'Instruct-uncensored-abliterated-18.4B-GGUF', please verify if 'abliterated' is the intended word. Possibly it should be 'obliterated'?
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 0% vs. threshold = 50% "Abliterated" is actually a real term in the AI safety/alignment community. It refers to a technique for removing certain behaviors or biases from language models through activation engineering. The term is intentionally different from "obliterated" and has a specific technical meaning. Given that this term appears consistently across many different model names in the log, it's clearly intentional. The comment is suggesting a correction that would actually be wrong. This is not about code changes - it's about a log file entry showing model names from HuggingFace. The comment is not about changes made in the diff, but rather questioning terminology in existing model names. Could "abliterated" be a legitimate variant spelling or technical jargon that I'm not aware of? However, given its consistent usage across multiple independent model repositories and authors, this strongly suggests it's intentional terminology rather than a widespread typo. The consistent usage across many different models from different authors/organizations (Huihui, gemma, Qwen, DavidAU, etc.) provides strong evidence this is intentional technical terminology. If it were a typo, it's unlikely to appear so consistently across unrelated projects. Additionally, this comment is about log output showing model names, not about code that needs to be changed. This comment should be deleted. "Abliterated" is legitimate technical terminology in the AI/ML community referring to a specific model modification technique. The comment incorrectly suggests it's a typo. Additionally, the comment is about model names in a log file, not about code changes that need to be made.

9. script_output.log:860

• Draft comment:
  Potential typographical error: The repository name "PleIAs/Monad" in line 860 looks suspicious. Please verify if "PleIAs" is spelled/capitalized correctly (e.g., it might be intended as "Pleias").
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

10. script_output.log:920

• Draft comment:
  Typo suggestion: The model name 'Huihui-Qwen3-VL-8B-Instruct-abliterated' seems unusual. Please verify if 'abliterated' is correct or if it should be spelled differently.
• Reason this comment was not posted:
  Comment was on unchanged code.

11. script_output.log:962

• Draft comment:
  Typo suggestion: The repository name 'google/gemma-3n-E2B-it-litert-lm' contains 'litert'. Please confirm if this is the intended spelling or if it should be something like 'literate' or 'libert'.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 0% vs. threshold = 50% This comment is incorrect. "LiteRT" (Lite Runtime) is Google's official name for their lightweight runtime framework (formerly TensorFlow Lite). The repository name is correct and intentional. The comment is asking the PR author to verify something that doesn't need verification - it's making an incorrect assumption that this is a typo when it's actually a legitimate product name. This type of comment wastes the PR author's time and should be deleted. Additionally, the comment is about a repository name that appears in log output, not about code that was changed in the diff. Could "litert" be an actual typo that happens to look like it could be a product name? However, given this is from Google's official organization and appears consistently in the logs, it's highly unlikely to be a typo. The comment is also not about any code change - it's about a repository name that appears in log output. Even if there was uncertainty about whether "litert" is correct, this comment is about a repository name in log output, not about code changes made in this PR. The PR author has no control over Google's repository naming. This comment should definitely be deleted. Delete this comment. "LiteRT" is Google's official product name (their lightweight runtime framework), not a typo. Additionally, this comment is about a repository name appearing in log output, not about any code changes in the diff, which violates the rule about only commenting on changes.

12. script_output.log:1566

• Draft comment:
  Typo: In the log message "Found 1 text models", the singular/plural form is mismatched. It should likely read "Found 1 text model" instead of "1 text models".
• Reason this comment was not posted:
  Comment was on unchanged code.

13. script_output.log:2755

• Draft comment:
  Typographical Note: The repository name 'kandinskylab' is in lowercase while similar entries (e.g. in 'Vortex5/Chaos-Unknown-12b') tend to use a capital letter. Please confirm if this lowercase usage is intentional or if it should be updated to 'Kandinskylab' for consistency.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 0% vs. threshold = 50% This comment is asking about the capitalization of a repository owner's username on Hugging Face. Repository owner names on Hugging Face are case-insensitive but have a canonical form. The actual username is "kandinskylab" (lowercase) as that's how it appears in the Hugging Face URLs. This is not something the PR author can change - it's the actual username of the repository owner. The comment is asking the author to "confirm if this lowercase usage is intentional" but the author has no control over how third-party repository owners name their accounts. This is not about code changes or anything actionable - it's about external data that the script is processing. The comment doesn't relate to any change made in the diff; it's just an observation about external repository naming. Could this comment be pointing out an actual inconsistency in how the code handles repository names? Maybe there's a normalization issue where some repos are being stored with different capitalizations? However, looking at the diff, these appear to be actual repository names from Hugging Face, not something the code is generating or modifying. The comment is about external repository naming on Hugging Face, which is outside the control of this PR. The repository owner's username is "kandinskylab" (lowercase) - that's the actual canonical name on Hugging Face. This isn't something that needs to be "confirmed" or "updated for consistency" because it's simply the correct name of the repository owner. The comment doesn't identify any code issue or actionable change. This comment should be deleted. It's asking about the capitalization of a third-party Hugging Face repository owner's username, which is not something the PR author can or should change. The lowercase "kandinskylab" is the actual correct username. This comment doesn't relate to any code changes and isn't actionable.

14. script_output.log:3086

• Draft comment:
  Minor grammatical suggestion: when printing a count of 1, consider using "model" instead of "models" (i.e. "Found 1 text model and 0 mmproj models") for clarity.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 0% vs. threshold = 50% This comment is about log output formatting - specifically pluralization. However, looking at the diff, this is a log OUTPUT file (script_output.log), not source code. The diff shows the results of running a script, not the script itself. Comments should be about code changes, not about the output of code. Additionally, this is a very minor stylistic suggestion about grammar in log messages. Even if this were about actual code changes, it would be borderline too minor to be useful. The comment doesn't point to any actual code that was changed - it's commenting on generated output. Could this be a legitimate suggestion if the underlying code that generates this output was changed in this PR? However, I don't see any code changes in the diff - only log output changes. The comment is about line 3086 in a log file, not source code. The diff only shows changes to script_output.log, which is a generated log file, not source code. Comments should be about code changes, not about the format of log output. This comment doesn't reference any actual code change and is purely about output formatting. Even if the underlying code were in the diff, this would be an extremely minor stylistic suggestion. This comment should be deleted. It's commenting on log output rather than code changes, and it's a very minor stylistic suggestion about pluralization that doesn't warrant a code change request.

Workflow ID: wflow_zkq4yKnEgb4naKRM

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}