knurlogic 0.1.0__tar.gz

knurlogic-0.1.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

knurlogic-0.1.0/NOTICE

@@ -0,0 +1,44 @@
+knurlogic
+Copyright 2026 Noah Zelezny
+
+This product is licensed under the Apache License, Version 2.0 (LICENSE).
+
+This product includes software developed by other projects, vendored under
+their own licenses. Each folder that carries vendored source records it in
+a THIRD-PARTY.md (license) and PROVENANCE.md (origin, version, digest).
+
+mlx-lm (https://github.com/ml-explore/mlx-lm)
+  MIT License, Copyright (c) 2023-2026 Apple Inc.
+  Model architectures in src/knurlogic/engine/families/qwen/architecture/
+  (qwen3_5.py, qwen3_5_moe.py; qwen4_exp.py via the mlx-lm fork at
+  https://github.com/eauchs/mlx-lm) and
+  src/knurlogic/engine/families/gemma4/architecture/ (gemma4.py,
+  gemma4_text.py).
+
+mlx-lm fork (the knurlogic author's own)
+  MIT License, Copyright (c) 2023-2025 Apple Inc.; changes by Noah Zelezny
+  src/knurlogic/engine/families/deepseek/architecture/deepseek_v4.py
+
+mlx-vlm (https://github.com/Blaizzy/mlx-vlm)
+  MIT License, Copyright (c) 2025 Prince Canuma
+  src/knurlogic/engine/families/glm5/architecture/glm5_next/ and the
+  vision towers in src/knurlogic/engine/families/{qwen,gemma4}/vision/.
+  hyper_connection.py (glm5_next/_mlx_vlm/models/deepseek_v4/):
+  MIT License, Copyright (c) 2026 Apple Inc.
+
+DeepSeek-V4 chat encoding
+(https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash)
+  MIT License, Copyright (c) 2023 DeepSeek
+  src/knurlogic/engine/templates/deepseek_v4.jinja (a Jinja port) and
+  tests/fixtures_deepseek_v4/ (the encoder and its golden outputs).
+
+vqlab (https://github.com/noahzelezny/VQLab)
+  Apache License 2.0, Copyright 2026 Noah Zelezny
+  src/knurlogic/engine/families/glm5/heads/glm5.py and the snapshot caches
+  in src/knurlogic/engine/mtp/caches.py.
+
+exo fork (https://github.com/noahzelezny/exo)
+  Apache License 2.0; the ported code is the knurlogic author's own
+  src/knurlogic/engine/mtp/batch_generator.py: ported from the maintainer's
+  own code in github.com/noahzelezny/exo (generator/mtp_batch_generate.py),
+  Apache-2.0. It contains no code from upstream exo.

knurlogic-0.1.0/PKG-INFO

@@ -0,0 +1,178 @@
+Metadata-Version: 2.4
+Name: knurlogic
+Version: 0.1.0
+Summary: Serve local language models on Apple Silicon, for people and agents.
+Author: Noah Zelezny
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/noahzelezny/Knurlogic
+Project-URL: Repository, https://github.com/noahzelezny/Knurlogic
+Project-URL: Issues, https://github.com/noahzelezny/Knurlogic/issues
+Project-URL: Changelog, https://github.com/noahzelezny/Knurlogic/blob/main/CHANGELOG.md
+Keywords: mlx,apple-silicon,llm,inference,openai,anthropic,mcp,local-models
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: MacOS X
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: MacOS
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+License-File: NOTICE
+Requires-Dist: mlx==0.31.2; platform_system == "Darwin"
+Requires-Dist: mlx-lm==0.31.3
+Requires-Dist: huggingface_hub>=0.24
+Requires-Dist: numpy>=1.24
+Requires-Dist: Pillow>=10.0
+Requires-Dist: rumps>=0.4; sys_platform == "darwin"
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: ruff; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Dynamic: license-file
+
+# Knurlogic
+
+Knurlogic serves local language models on Apple Silicon. It works out the
+settings a model needs and whether it fits in memory before loading it,
+then runs it behind an OpenAI- and Anthropic-compatible server. People use
+it through a web page; agents use it through an MCP server. One model can
+also be split across two or more Macs.
+
+## Requirements
+
+* A Mac with Apple Silicon and enough unified memory for the model.
+* Python 3.10 or newer.
+* mlx 0.31.2 and mlx-lm 0.31.3, pinned exactly (pip installs them).
+
+## Install
+
+    pip install knurlogic
+
+From source:
+
+    git clone https://github.com/noahzelezny/Knurlogic
+    cd Knurlogic && pip install -e .
+
+## Quickstart
+
+Get a model in MLX format: search and download it from the page's model
+picker (Hugging Face), or with the `hf` CLI that comes with
+`huggingface_hub`:
+
+    pip install -U huggingface_hub
+    hf download mlx-community/gemma-4-e4b-it-8bit \
+      --local-dir ~/Knurlogic/Models/gemma-4-e4b-it-8bit
+
+`knurlogic models` lists the models already on this Mac (Knurlogic's own
+folder, the Hugging Face cache, LM Studio, Ollama and exo folders).
+
+Check it and serve it:
+
+    knurlogic doctor ~/Knurlogic/Models/gemma-4-e4b-it-8bit
+    knurlogic serve  ~/Knurlogic/Models/gemma-4-e4b-it-8bit
+
+Then ask it something:
+
+    curl http://127.0.0.1:8080/v1/chat/completions \
+      -H 'Content-Type: application/json' \
+      -d '{"model": "local", "messages": [{"role": "user", "content": "Hello"}]}'
+
+The page is at http://127.0.0.1:8080/: what is loaded, where the memory
+went, a chat and the settings. `knurlogic ui` opens the page at
+http://127.0.0.1:8899/ without loading anything, with a Launch button for
+every model on the disk, and a gear in the macOS menu bar (`--no-menubar`
+skips it) that shows what is loaded and opens or quits the page.
+
+## Connect a harness
+
+`knurlogic connect --port 8080` prints these for a running server.
+
+**Claude Code** (or any Anthropic Messages client):
+
+    ANTHROPIC_BASE_URL=http://127.0.0.1:8080 \
+    ANTHROPIC_API_KEY=x \
+    ANTHROPIC_DEFAULT_OPUS_MODEL=local \
+    ANTHROPIC_DEFAULT_SONNET_MODEL=local \
+    ANTHROPIC_DEFAULT_HAIKU_MODEL=local \
+    API_TIMEOUT_MS=3000000 \
+    claude
+
+**OpenAI-compatible clients** (Zed, Cline, Continue, Open WebUI): base URL
+`http://127.0.0.1:8080/v1`, any API key, model `local`.
+
+**Responses-API clients** (the OpenAI SDK's `client.responses`, Codex-style
+tools): the same base URL; `POST /v1/responses` with function tools,
+streaming and reasoning summaries. `previous_response_id` and `store: true`
+are refused: the server keeps no conversation, so resend the input.
+
+**Ollama clients** (Open WebUI's Ollama mode, Continue, the `ollama`
+libraries): `OLLAMA_HOST=http://127.0.0.1:8080`. `/api/chat`, `/api/generate`
+(streaming NDJSON by default), `/api/tags`, `/api/show` and `/api/version`
+are served, with images in chat messages on a vision model.
+
+**MCP**, for an agent that should manage models rather than talk to one:
+
+    claude mcp add knurlogic -- knurlogic mcp
+
+Its tools are `models`, `fit`, `settings`, `drafting`, `ready`, `load`,
+`state`, `unload` and `deps`; `knurlogic mcp --list` describes each.
+`load` refuses a model that does not fit.
+
+## Two or more Macs
+
+To split one model across two or more Macs (up to 16) joined by Thunderbolt,
+install the same
+knurlogic version and the model on both, and run on each:
+
+    knurlogic ui --host cluster
+
+The Macs find each other over Bonjour (`--peer HOST` names one directly, on each machine
+(or link them with Thunderbolt);
+`knurlogic doctor --cluster` says what is in the way). Then launch the
+model from the page with the machines selected, or with the MCP `load`
+tool's `machines` argument. The TCP ring works for any number of Macs;
+RDMA (jaccl) needs every pair cabled with Thunderbolt 5, and beyond two
+Macs it is experimental and untested.
+
+## Settings
+
+Every setting is resolved from the model's `config.json` and the memory
+available, and shown in the page's Settings panel and at `/settings.json`.
+Two presets cover most needs: `default` (fastest) and `lean` (more
+context in less memory: 512-token prompt chunks, MTP off, 8-bit KV).
+`serve --tune default|lean` picks one, `--kv-bits 8` stores the KV cache
+in about half the memory, and `--set KEY=VALUE` overrides any single
+setting. A context length past the model's native window turns on YaRN
+for the Qwen families (up to 1,048,576 tokens).
+
+## Supported models
+
+| Family | `model_type` | Vision | Drafting (MTP) | 8-bit KV |
+|---|---|---|---|---|
+| Qwen 3.5 / 3.6 | `qwen3_5`, `qwen3_5_moe` | yes | if the model has a head | yes |
+| Qwen 3.8 | `qwen4_exp` | yes | if the model has a head | yes |
+| Gemma 4 | `gemma4`, `gemma4_text` | yes | no | yes |
+| GLM-5 | `glm5_next` | yes | if the model has a head | yes |
+| DeepSeek-V4 | `deepseek_v4` | no | no | no |
+
+VQ-quantized models (published with their own `model.py`) are supported
+for these families; each runs the `model.py` it ships. The page tags a
+Hugging Face model "update" when the Hub has a newer revision (checked once
+at start; `knurlogic ui --offline` or `HF_HUB_OFFLINE=1` skips it).
+
+## Code
+
+How the package is laid out and who may import whom: [docs/architecture.md](docs/architecture.md).
+
+## Status
+
+Alpha. Expect rough edges and breaking changes before 1.0; see
+CHANGELOG.md.
+
+## License
+
+Apache-2.0. Vendored third-party code is listed in NOTICE.

knurlogic-0.1.0/README.md

@@ -0,0 +1,142 @@
+# Knurlogic
+
+Knurlogic serves local language models on Apple Silicon. It works out the
+settings a model needs and whether it fits in memory before loading it,
+then runs it behind an OpenAI- and Anthropic-compatible server. People use
+it through a web page; agents use it through an MCP server. One model can
+also be split across two or more Macs.
+
+## Requirements
+
+* A Mac with Apple Silicon and enough unified memory for the model.
+* Python 3.10 or newer.
+* mlx 0.31.2 and mlx-lm 0.31.3, pinned exactly (pip installs them).
+
+## Install
+
+    pip install knurlogic
+
+From source:
+
+    git clone https://github.com/noahzelezny/Knurlogic
+    cd Knurlogic && pip install -e .
+
+## Quickstart
+
+Get a model in MLX format: search and download it from the page's model
+picker (Hugging Face), or with the `hf` CLI that comes with
+`huggingface_hub`:
+
+    pip install -U huggingface_hub
+    hf download mlx-community/gemma-4-e4b-it-8bit \
+      --local-dir ~/Knurlogic/Models/gemma-4-e4b-it-8bit
+
+`knurlogic models` lists the models already on this Mac (Knurlogic's own
+folder, the Hugging Face cache, LM Studio, Ollama and exo folders).
+
+Check it and serve it:
+
+    knurlogic doctor ~/Knurlogic/Models/gemma-4-e4b-it-8bit
+    knurlogic serve  ~/Knurlogic/Models/gemma-4-e4b-it-8bit
+
+Then ask it something:
+
+    curl http://127.0.0.1:8080/v1/chat/completions \
+      -H 'Content-Type: application/json' \
+      -d '{"model": "local", "messages": [{"role": "user", "content": "Hello"}]}'
+
+The page is at http://127.0.0.1:8080/: what is loaded, where the memory
+went, a chat and the settings. `knurlogic ui` opens the page at
+http://127.0.0.1:8899/ without loading anything, with a Launch button for
+every model on the disk, and a gear in the macOS menu bar (`--no-menubar`
+skips it) that shows what is loaded and opens or quits the page.
+
+## Connect a harness
+
+`knurlogic connect --port 8080` prints these for a running server.
+
+**Claude Code** (or any Anthropic Messages client):
+
+    ANTHROPIC_BASE_URL=http://127.0.0.1:8080 \
+    ANTHROPIC_API_KEY=x \
+    ANTHROPIC_DEFAULT_OPUS_MODEL=local \
+    ANTHROPIC_DEFAULT_SONNET_MODEL=local \
+    ANTHROPIC_DEFAULT_HAIKU_MODEL=local \
+    API_TIMEOUT_MS=3000000 \
+    claude
+
+**OpenAI-compatible clients** (Zed, Cline, Continue, Open WebUI): base URL
+`http://127.0.0.1:8080/v1`, any API key, model `local`.
+
+**Responses-API clients** (the OpenAI SDK's `client.responses`, Codex-style
+tools): the same base URL; `POST /v1/responses` with function tools,
+streaming and reasoning summaries. `previous_response_id` and `store: true`
+are refused: the server keeps no conversation, so resend the input.
+
+**Ollama clients** (Open WebUI's Ollama mode, Continue, the `ollama`
+libraries): `OLLAMA_HOST=http://127.0.0.1:8080`. `/api/chat`, `/api/generate`
+(streaming NDJSON by default), `/api/tags`, `/api/show` and `/api/version`
+are served, with images in chat messages on a vision model.
+
+**MCP**, for an agent that should manage models rather than talk to one:
+
+    claude mcp add knurlogic -- knurlogic mcp
+
+Its tools are `models`, `fit`, `settings`, `drafting`, `ready`, `load`,
+`state`, `unload` and `deps`; `knurlogic mcp --list` describes each.
+`load` refuses a model that does not fit.
+
+## Two or more Macs
+
+To split one model across two or more Macs (up to 16) joined by Thunderbolt,
+install the same
+knurlogic version and the model on both, and run on each:
+
+    knurlogic ui --host cluster
+
+The Macs find each other over Bonjour (`--peer HOST` names one directly, on each machine
+(or link them with Thunderbolt);
+`knurlogic doctor --cluster` says what is in the way). Then launch the
+model from the page with the machines selected, or with the MCP `load`
+tool's `machines` argument. The TCP ring works for any number of Macs;
+RDMA (jaccl) needs every pair cabled with Thunderbolt 5, and beyond two
+Macs it is experimental and untested.
+
+## Settings
+
+Every setting is resolved from the model's `config.json` and the memory
+available, and shown in the page's Settings panel and at `/settings.json`.
+Two presets cover most needs: `default` (fastest) and `lean` (more
+context in less memory: 512-token prompt chunks, MTP off, 8-bit KV).
+`serve --tune default|lean` picks one, `--kv-bits 8` stores the KV cache
+in about half the memory, and `--set KEY=VALUE` overrides any single
+setting. A context length past the model's native window turns on YaRN
+for the Qwen families (up to 1,048,576 tokens).
+
+## Supported models
+
+| Family | `model_type` | Vision | Drafting (MTP) | 8-bit KV |
+|---|---|---|---|---|
+| Qwen 3.5 / 3.6 | `qwen3_5`, `qwen3_5_moe` | yes | if the model has a head | yes |
+| Qwen 3.8 | `qwen4_exp` | yes | if the model has a head | yes |
+| Gemma 4 | `gemma4`, `gemma4_text` | yes | no | yes |
+| GLM-5 | `glm5_next` | yes | if the model has a head | yes |
+| DeepSeek-V4 | `deepseek_v4` | no | no | no |
+
+VQ-quantized models (published with their own `model.py`) are supported
+for these families; each runs the `model.py` it ships. The page tags a
+Hugging Face model "update" when the Hub has a newer revision (checked once
+at start; `knurlogic ui --offline` or `HF_HUB_OFFLINE=1` skips it).
+
+## Code
+
+How the package is laid out and who may import whom: [docs/architecture.md](docs/architecture.md).
+
+## Status
+
+Alpha. Expect rough edges and breaking changes before 1.0; see
+CHANGELOG.md.
+
+## License
+
+Apache-2.0. Vendored third-party code is listed in NOTICE.