Skip to content

Actions: AtomicBot-ai/mlx-vlm

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
20 workflow runs
20 workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
feat: extend multimodal weight identification in model loading Build & Release MLX-VLM Server (macOS ARM64) #20: Commit d784fc5 pushed by Vect0rM
5m 18s main
main
5m 18s
feat: enhance model loading with text-only and multimodal support Build & Release MLX-VLM Server (macOS ARM64) #19: Commit 9dfe03d pushed by Vect0rM
4m 18s main
main
4m 18s
fix: pin transformers==5.9.0 to keep Gemma4 processor torch-free Build & Release MLX-VLM Server (macOS ARM64) #18: Commit 70fef26 pushed by Vect0rM
4m 25s main
main
4m 25s
fix: require transformers>=5.9.0 to keep Gemma4 processor torch-free Build & Release MLX-VLM Server (macOS ARM64) #17: Commit d744967 pushed by Vect0rM
57s main
main
57s
fix: drop vestigial KV-shared weights for MLX-format gemma4 checkpoints Build & Release MLX-VLM Server (macOS ARM64) #16: Commit 19e0496 pushed by Vect0rM
4m 24s main
main
4m 24s
merge: sync upstream Blaizzy/mlx-vlm@main (v0.6.3) Build & Release MLX-VLM Server (macOS ARM64) #15: Commit 8ec334a pushed by Vect0rM
4m 34s main
main
4m 34s
fix: use gemma4_unified <|turn> template + stop on <turn|> Build & Release MLX-VLM Server (macOS ARM64) #14: Commit 0a82347 pushed by Vect0rM
4m 39s main
main
4m 39s
fix: honor suppress_tokens and install Gemma chat-template fallback Build & Release MLX-VLM Server (macOS ARM64) #13: Commit 654a520 pushed by Vect0rM
4m 28s main
main
4m 28s
Adjust Gemma4 quantization predicate (#1288) Build & Release MLX-VLM Server (macOS ARM64) #12: Commit 88d260c pushed by Vect0rM
5m 9s main
main
5m 9s
Add video input support for Gemma 4 12B. (#1292) Build & Release MLX-VLM Server (macOS ARM64) #11: Commit f42f567 pushed by Vect0rM
5m 40s main
main
5m 40s
Enhance accepted counts handling in LanguageModel Build & Release MLX-VLM Server (macOS ARM64) #10: Commit aed5482 pushed by Vect0rM
5m 22s main
main
5m 22s
Enhance generation arguments and EOS token handling Build & Release MLX-VLM Server (macOS ARM64) #9: Commit a7aa668 pushed by Vect0rM
4m 56s main
main
4m 56s
Add final SSE chunk for OpenAI-compatible clients to report accurate … Build & Release MLX-VLM Server (macOS ARM64) #8: Commit 7789b0a pushed by Vect0rM
5m 19s main
main
5m 19s
merge: sync upstream Blaizzy/mlx-vlm@main (incl. PR #1139 MTP SWA mas… Build & Release MLX-VLM Server (macOS ARM64) #7: Commit 3ec92c9 pushed by Vect0rM
4m 26s main
main
4m 26s
Refactor sampling logic in ResponseGenerator to enforce lossless targ… Build & Release MLX-VLM Server (macOS ARM64) #6: Commit bf1bb38 pushed by Vect0rM
5m 46s main
main
5m 46s
Add EOS token aggregation in server.py Build & Release MLX-VLM Server (macOS ARM64) #5: Commit f45d305 pushed by Vect0rM
5m 18s main
main
5m 18s
Enhance top_p_sampling function in sample_utils.py to support arbitra… Build & Release MLX-VLM Server (macOS ARM64) #4: Commit f74bd14 pushed by Vect0rM
4m 15s main
main
4m 15s
Enhance chat template handling in server.py by introducing support fo… Build & Release MLX-VLM Server (macOS ARM64) #3: Commit 10150b0 pushed by Vect0rM
5m 11s main
main
5m 11s
Implement single-model mode handling in server.py to prevent cache un… Build & Release MLX-VLM Server (macOS ARM64) #2: Commit 0eb7e32 pushed by Vect0rM
4m 55s main
main
4m 55s
Add llguidance and mlx-audio packages to uv.lock; remove GitHub workf… Build & Release MLX-VLM Server (macOS ARM64) #1: Commit 0c4c1c4 pushed by Vect0rM
4m 42s main
main
4m 42s