Skip to content

Commit 2255239

Browse files
aittalamaittalam
committed
Merge branch 'main' into llamacpp-7b8443a
2 parents 45bbf99 + edb2fd0 commit 2255239

11 files changed

Lines changed: 357 additions & 97 deletions

File tree

‎.github/pull_request_template.md‎

Lines changed: 39 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,39 @@
1+
## Description
2+
<!-- What does this PR do? -->
3+
4+
5+
## PR Type
6+
<!-- Delete the types that don't apply -->
7+
8+
- 🆕 New Feature
9+
- 🐛 Bug Fix
10+
- 💅 Refactor
11+
- 📚 Documentation
12+
- 🚦 Infrastructure
13+
14+
## Relevant issues
15+
<!-- e.g. "Fixes #123" -->
16+
17+
## Checklist
18+
<!-- If this checklist is deleted from the PR submission it may be closed -->
19+
- [ ] I understand the code I am submitting.
20+
- [ ] I have run this code locally and verified the change.
21+
- [ ] New and existing tests pass locally, or I have explained why tests were not run.
22+
- [ ] Documentation was updated where necessary.
23+
- [ ] If I changed code in `llama.cpp/`, `whisper.cpp/`, or `stable-diffusion.cpp/`, I also updated the matching `*.patches/` files.
24+
- [ ] I have read and followed the [contribution guidelines](https://github.com/mozilla-ai/llamafile/blob/main/CONTRIBUTING.md).
25+
- [ ] **AI Usage:**
26+
- [ ] No AI was used.
27+
- [ ] AI was used in an assistive capacity.
28+
- [ ] This PR includes substantial AI-generated content.
29+
30+
## AI Usage Information
31+
<!-- Optional: if AI was used, briefly describe how -->
32+
33+
- AI Model used:
34+
- AI Developer Tool used:
35+
- Any other info you'd like to share:
36+
37+
When answering reviewer questions, please respond yourself rather than pasting reviewer comments into an AI system and posting the reply back unchanged.
38+
39+
- [ ] I am an AI Agent filling out this form (check box if true)

‎CONTRIBUTING.md‎

Lines changed: 176 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,176 @@
1+
# Contributing to llamafile
2+
3+
Thank you for your interest in contributing to llamafile.
4+
5+
We welcome fixes, docs improvements, tests, build work, and larger feature work.
6+
7+
Submodule changes (`llama.cpp/`, `whisper.cpp/`, `stable-diffusion.cpp/`) are applied as patches rather than committed directly. If your change should also go upstream, open a PR to the upstream repository (e.g., [llama.cpp](https://github.com/ggml-org/llama.cpp)). Otherwise, follow the [submodule changes workflow](#submodule-changes) described below.
8+
9+
## Before You Start
10+
11+
### Check for duplicates
12+
13+
Before starting new work:
14+
15+
- Search [existing issues](https://github.com/mozilla-ai/llamafile/issues) for duplicates
16+
- Check [open pull requests](https://github.com/mozilla-ai/llamafile/pulls) to see if someone is already working on it
17+
- For bugs, verify the issue still exists on `main`
18+
19+
### Discuss major changes first
20+
21+
Please open an issue before starting larger changes such as:
22+
23+
- new user-facing features
24+
- architectural changes
25+
- changes to public behavior or defaults
26+
- new dependencies
27+
- significant build or packaging changes
28+
29+
This helps us stay aligned and avoids duplicate work.
30+
31+
## Development Setup
32+
33+
### Prerequisites
34+
35+
You will need:
36+
37+
- GNU `make` (called `gmake` on some systems)
38+
- `sha256sum` or a working `cc`
39+
- `wget` or `curl`
40+
- `unzip`
41+
- Git
42+
43+
Windows contributors can use [MSYS2](https://www.msys2.org/) or WSL. See [docs/building_dlls.md](docs/building_dlls.md) for detailed Windows setup instructions.
44+
45+
### Quick Start
46+
47+
```sh
48+
# 1. Fork the repository on GitHub
49+
50+
# 2. Clone your fork
51+
git clone https://github.com/YOUR_USERNAME/llamafile.git
52+
cd llamafile
53+
54+
# 3. Add upstream remote
55+
git remote add upstream https://github.com/mozilla-ai/llamafile.git
56+
57+
# 4. Set up submodules, patches, and toolchain
58+
make setup
59+
60+
# 5. Build with cosmocc's make
61+
.cosmocc/4.0.2/bin/make -j8
62+
63+
# 6. Run the default test suite
64+
.cosmocc/4.0.2/bin/make check
65+
```
66+
67+
`make setup` initializes submodules, applies llamafile-specific patches, and downloads the `cosmocc` toolchain into `.cosmocc/`.
68+
69+
For builds and tests, use `.cosmocc/4.0.2/bin/make`, not your system `make`.
70+
71+
## Making Changes
72+
73+
### 1. Create a branch
74+
75+
Always work on a branch, not directly on `main`:
76+
77+
```sh
78+
git checkout -b docs/your-change
79+
```
80+
81+
Common branch prefixes:
82+
83+
- `docs/` for documentation
84+
- `fix/` for bug fixes
85+
- `feature/` for new features
86+
- `build/` for build and tooling changes
87+
88+
### 2. Make your changes
89+
90+
There are two common workflows in this repo.
91+
92+
#### Core code changes
93+
94+
For changes in directories like:
95+
96+
- `llamafile/`
97+
- `whisperfile/`
98+
- `docs/`
99+
- `tests/`
100+
101+
you can edit files normally, rebuild, test, and commit as usual.
102+
103+
#### Submodule changes
104+
105+
The following directories are submodules:
106+
107+
- `llama.cpp/`
108+
- `whisper.cpp/`
109+
- `stable-diffusion.cpp/`
110+
111+
If you change code inside one of those directories, you also need to save those changes as patches in the matching `*.patches/` directory.
112+
113+
When working inside a submodule, follow that submodule's local coding and contribution guidelines in addition to this repository's workflow.
114+
115+
Example for `llama.cpp`:
116+
117+
```sh
118+
cd llama.cpp
119+
../tools/generate-patches.sh --output-dir ../llama.cpp.patches
120+
```
121+
122+
After generating patches, verify them from a clean state:
123+
124+
```sh
125+
make reset-repo
126+
make setup
127+
.cosmocc/4.0.2/bin/make -j8
128+
.cosmocc/4.0.2/bin/make check
129+
```
130+
131+
For a more detailed walkthrough of the patch-based workflow, see [docs/skills/llamafile/development.md](docs/skills/llamafile/development.md#making-changes-to-a-submodule).
132+
133+
### 3. Write tests
134+
135+
Please add or update tests whenever your change affects behavior.
136+
137+
- New features should include tests
138+
- Bug fixes should include a regression test when practical
139+
- Docs-only changes usually do not need tests
140+
- Avoid mixing unrelated changes in one pull request
141+
142+
There are also integration tests under [tests/integration/README.md](tests/integration/README.md) if you want to validate changes with a real model.
143+
144+
### 4. Update documentation
145+
146+
If your change affects how developers or users work with llamafile, update the relevant docs in `README.md` or `docs/`.
147+
148+
If you add a new page to `docs/`, also add it to [`docs/SUMMARY.md`](docs/SUMMARY.md) — that file controls the GitBook navigation and is maintained by hand. CI will catch any SUMMARY entries that point to missing files, but it will not catch a new file that was never added to SUMMARY.
149+
150+
### 5. Commit your changes
151+
152+
Use clear commit messages:
153+
154+
```sh
155+
git commit -m "Fix server startup when model path is missing"
156+
git commit -m "Update contributor guide for patch workflow"
157+
```
158+
159+
## Submitting Changes
160+
161+
Before opening a pull request, please make sure:
162+
163+
- the project builds cleanly
164+
- the default test suite passes
165+
- submodule changes have been converted into patch files
166+
- related documentation has been updated
167+
- the change is focused and easy to review
168+
- you are ready to explain and maintain the code you changed
169+
170+
## Useful Docs
171+
172+
- [README.md](README.md)
173+
- [docs/source_installation.md](docs/source_installation.md)
174+
- [docs/running_llamafile.md](docs/running_llamafile.md)
175+
- [docs/creating_llamafiles.md](docs/creating_llamafiles.md)
176+
- [tests/integration/README.md](tests/integration/README.md)

‎README.md‎

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -21,18 +21,18 @@ framework that collapses all the complexity of LLMs down to
2121
a single-file executable (called a "llamafile") that runs
2222
locally on most operating systems and CPU archiectures, with no installation.
2323

24-
llamafile also includes **[whisperfile](docs/whisperfile/index.md)**, a single-file speech-to-text tool built on [whisper.cpp](https://github.com/ggerganov/whisper.cpp) and the same Cosmopolitan packaging. It supports transcription and translation of audio files across all the same platforms, with no installation required.
24+
llamafile also includes **[whisperfile](https://docs.mozilla.ai/llamafile/whisperfile)**, a single-file speech-to-text tool built on [whisper.cpp](https://github.com/ggerganov/whisper.cpp) and the same Cosmopolitan packaging. It supports transcription and translation of audio files across all the same platforms, with no installation required.
2525

2626

27-
## v0.10.0
27+
## v0.10.*
2828

2929
**llamafile versions starting from 0.10.0 use a new build system**, aimed at keeping our code more easily
3030
aligned with the latest versions of llama.cpp. This means they support more recent models and functionalities,
3131
but at the same time they might be missing some of
3232
the features you were accustomed to (check out [this doc](README_0.10.0.md) for a high-level description of what has been done). If you liked
3333
the "classic experience" more, you will always be able to access the previous versions from our
3434
[releases](https://github.com/mozilla-ai/llamafile/releases) page. Our pre-built llamafiles always
35-
show which version of the server they have been bundled with ([0.9.* example](https://huggingface.co/mozilla-ai/llava-v1.5-7b-llamafile), [0.10.* example](https://huggingface.co/mozilla-ai/llamafile_0.10.0)), so you will always know
35+
show which version of the server they have been bundled with ([0.9.* example](https://huggingface.co/mozilla-ai/llava-v1.5-7b-llamafile), [0.10.* example](https://huggingface.co/mozilla-ai/llamafile_0.10)), so you will always know
3636
which version of the software you are downloading.
3737

3838

@@ -47,7 +47,7 @@ Download and run your first llamafile in minutes:
4747

4848
```sh
4949
# Download an example model (Qwen3.5 0.8B)
50-
curl -LO https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-0.8B-Q8_0.llamafile
50+
curl -LO https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Qwen3.5-0.8B-Q8_0.llamafile
5151

5252
# Make it executable (macOS/Linux/BSD)
5353
chmod +x Qwen3.5-0.8B-Q8_0.llamafile
@@ -58,25 +58,25 @@ chmod +x Qwen3.5-0.8B-Q8_0.llamafile
5858

5959
We chose this model because that's the smallest one we have
6060
built a llamafile for, so most likely to work out-of-the-box for you.
61-
If you have powerful hardware and/or GPUs, [feel free to choose](docs/example_llamafiles.md)
61+
If you have powerful hardware and/or GPUs, [feel free to choose](https://docs.mozilla.ai/llamafile/getting-started/example_llamafiles)
6262
larger and more expressive models which should provide more accurate
6363
responses.
6464

6565
**Windows users:** Rename the file to add `.exe` extension before running.
6666

6767
## Documentation
6868

69-
Check the full documentation in the [docs/](docs/) folder, or directly jump into one of the following subsections:
70-
71-
- [Quickstart](docs/quickstart.md)
72-
- [Example llamafiles](docs/example_llamafiles.md)
73-
- [Running a llamafile](docs/running_llamafile.md)
74-
- [Creating llamafiles](docs/creating_llamafiles.md)
75-
- [Source installation](docs/source_installation.md)
76-
- [Technical details](docs/technical_details.md)
77-
- [Supported Systems](docs/support.md)
78-
- [Troubleshooting](docs/troubleshooting.md)
79-
- [Whisperfile](docs/whisperfile/index.md)
69+
Check the full documentation at [docs.mozilla.ai/llamafile](https://docs.mozilla.ai/llamafile), or directly jump into one of the following subsections:
70+
71+
- [Quickstart](https://docs.mozilla.ai/llamafile/getting-started/quickstart)
72+
- [Example llamafiles](https://docs.mozilla.ai/llamafile/getting-started/example_llamafiles)
73+
- [Running a llamafile](https://docs.mozilla.ai/llamafile/using-llamafile/running_llamafile)
74+
- [Creating llamafiles](https://docs.mozilla.ai/llamafile/using-llamafile/creating_llamafiles)
75+
- [Source installation](https://docs.mozilla.ai/llamafile/using-llamafile/source_installation)
76+
- [Technical details](https://docs.mozilla.ai/llamafile/reference/technical_details)
77+
- [Supported Systems](https://docs.mozilla.ai/llamafile/reference/support)
78+
- [Troubleshooting](https://docs.mozilla.ai/llamafile/reference/troubleshooting)
79+
- [Whisperfile](https://docs.mozilla.ai/llamafile/whisperfile)
8080

8181

8282
## Licensing

‎README_0.10.0.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -53,7 +53,7 @@ mode) are new.
5353
[20251218](https://github.com/mozilla-ai/llamafile/discussions/845)
5454
- added Metal support: GPU on MacOS ARM64 is supported by compiling a small module
5555
using the Xcode Command Line Tools, which need to be installed. Check our docs at
56-
[docs/support.md#gpu-support](docs/support.md#gpu-support) for more info.
56+
[our support docs](https://docs.mozilla.ai/llamafile/reference/support#gpu-support) for more info.
5757
- Metal works both in llamafile (called either as TUI or with the --server flag)
5858
and in llama-server.
5959

‎docs/example_llamafiles.md‎

Lines changed: 14 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -1,24 +1,24 @@
11
We provide example llamafiles for a variety of models, so you can easily try out llamafile
22
with different kinds of LLMs. The following table lists llamafiles bundled with the latest
3-
available version of the server (v0.10.0). The smaller the file is, the more easily it will
3+
available version of the server (v0.10.*). The smaller the file is, the more easily it will
44
run on your computer, even if no GPU is present (as a reference, Qwen3.5 0.8B Q8 generates
55
text on a Raspberry Pi5 at ~8 tokens/sec).
66

77
| Model | Size | License | llamafile |
88
| --- | --- | --- | --- |
9-
| [Qwen3.5 0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B) Q8_0 | 1.6 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-0.8B-Q8_0.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-0.8B-Q8_0.llamafile) |
10-
| [Qwen3.5 2B](https://huggingface.co/Qwen/Qwen3.5-2B) Q8_0 | 3.2 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-2B-Q8_0.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-2B-Q8_0.llamafile) |
11-
| [Ministral 3 3B Instruct 2512](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) Q4_K_M | 3.4 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Ministral-3-3B-Instruct-2512-Q4_K_M.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Ministral-3-3B-Instruct-2512-Q4_K_M.llamafile) |
12-
| [Qwen3.5 4B](https://huggingface.co/Qwen/Qwen3.5-4B) Q5_K_S | 4.1 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-4B-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-4B-Q5_K_S.llamafile) |
13-
| [llava v1.6 mistral 7b](https://huggingface.co/liuhaotian/llava-v1.6-mistral-7b) Q4_K_M | 5.3 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [llava-v1.6-mistral-7b-Q4_K_M.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/llava-v1.6-mistral-7b-Q4_K_M.llamafile) |
14-
| [Apertus 8B Instruct 2509](https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509) | 5.9 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Apertus-8B-Instruct-2509.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Apertus-8B-Instruct-2509.llamafile) |
15-
| [Qwen3.5 9B](https://huggingface.co/Qwen/Qwen3.5-9B) Q5_K_S | 7.4 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-9B-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-9B-Q5_K_S.llamafile) |
16-
| [Ministral 3 3B Instruct 2512](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) BF16 | 7.8 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Ministral-3-3B-Instruct-2512-BF16.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Ministral-3-3B-Instruct-2512-BF16.llamafile) |
17-
| [llava v1.6 mistral 7b](https://huggingface.co/liuhaotian/llava-v1.6-mistral-7b) Q8_0 | 8.4 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [llava-v1.6-mistral-7b-Q8_0.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/llava-v1.6-mistral-7b-Q8_0.llamafile) |
18-
| [gpt-oss 20b](https://huggingface.co/openai/gpt-oss-20b) mxfp4 | 12 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [gpt-oss-20b-mxfp4.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/gpt-oss-20b-mxfp4.llamafile) |
19-
| [gpt-oss 20b](https://huggingface.co/openai/gpt-oss-20b) Q5_K_S | 12 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [gpt-oss-20b-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/gpt-oss-20b-Q5_K_S.llamafile) |
20-
| [LFM2 24B A2B](https://huggingface.co/LiquidAI/LFM2-24B-A2B) Q5_K_M | 16 GB | [lfm1.0](https://huggingface.co/LiquidAI/LFM2-24B-A2B/blob/main/LICENSE) | [LFM2-24B-A2B-Q5_K_M.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/LFM2-24B-A2B-Q5_K_M.llamafile) |
21-
| [Qwen3.5 27B](https://huggingface.co/Qwen/Qwen3.5-27B) Q5_K_S | 19 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-27B-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-27B-Q5_K_S.llamafile) |
9+
| [Qwen3.5 0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B) Q8_0 | 1.6 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-0.8B-Q8_0.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Qwen3.5-0.8B-Q8_0.llamafile) |
10+
| [Qwen3.5 2B](https://huggingface.co/Qwen/Qwen3.5-2B) Q8_0 | 3.2 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-2B-Q8_0.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Qwen3.5-2B-Q8_0.llamafile) |
11+
| [Ministral 3 3B Instruct 2512](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) Q4_K_M | 3.4 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Ministral-3-3B-Instruct-2512-Q4_K_M.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Ministral-3-3B-Instruct-2512-Q4_K_M.llamafile) |
12+
| [Qwen3.5 4B](https://huggingface.co/Qwen/Qwen3.5-4B) Q5_K_S | 4.1 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-4B-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Qwen3.5-4B-Q5_K_S.llamafile) |
13+
| [llava v1.6 mistral 7b](https://huggingface.co/liuhaotian/llava-v1.6-mistral-7b) Q4_K_M | 5.3 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [llava-v1.6-mistral-7b-Q4_K_M.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/llava-v1.6-mistral-7b-Q4_K_M.llamafile) |
14+
| [Apertus 8B Instruct 2509](https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509) | 5.9 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Apertus-8B-Instruct-2509.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Apertus-8B-Instruct-2509.llamafile) |
15+
| [Qwen3.5 9B](https://huggingface.co/Qwen/Qwen3.5-9B) Q5_K_S | 7.4 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-9B-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Qwen3.5-9B-Q5_K_S.llamafile) |
16+
| [Ministral 3 3B Instruct 2512](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) BF16 | 7.8 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Ministral-3-3B-Instruct-2512-BF16.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Ministral-3-3B-Instruct-2512-BF16.llamafile) |
17+
| [llava v1.6 mistral 7b](https://huggingface.co/liuhaotian/llava-v1.6-mistral-7b) Q8_0 | 8.4 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [llava-v1.6-mistral-7b-Q8_0.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/llava-v1.6-mistral-7b-Q8_0.llamafile) |
18+
| [gpt-oss 20b](https://huggingface.co/openai/gpt-oss-20b) mxfp4 | 12 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [gpt-oss-20b-mxfp4.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/gpt-oss-20b-mxfp4.llamafile) |
19+
| [gpt-oss 20b](https://huggingface.co/openai/gpt-oss-20b) Q5_K_S | 12 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [gpt-oss-20b-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/gpt-oss-20b-Q5_K_S.llamafile) |
20+
| [LFM2 24B A2B](https://huggingface.co/LiquidAI/LFM2-24B-A2B) Q5_K_M | 16 GB | [lfm1.0](https://huggingface.co/LiquidAI/LFM2-24B-A2B/blob/main/LICENSE) | [LFM2-24B-A2B-Q5_K_M.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/LFM2-24B-A2B-Q5_K_M.llamafile) |
21+
| [Qwen3.5 27B](https://huggingface.co/Qwen/Qwen3.5-27B) Q5_K_S | 19 GB | [Apache 2.0](https://choosealicense.com/licenses/apache-2.0/) | [Qwen3.5-27B-Q5_K_S.llamafile](https://huggingface.co/mozilla-ai/llamafile_0.10/resolve/main/Qwen3.5-27B-Q5_K_S.llamafile) |
2222

2323
## Legacy llamafiles
2424

0 commit comments

Comments
 (0)