Spaces:
Sleeping
Sleeping
Commit History
-
Revert to stable config: BATCH_TOKENS=512, CONCURRENT=8
Verified signatureb593f27 -
Reduce MAX_BATCH_TOKENS for CPU stability
Verified signature60f06b1 -
MAX_BATCH_TOKENS=32768, MAX_INPUT_LENGTH=8192
Verified signature11bc8de -
Increase MAX_BATCH_TOKENS to 32768, MAX_INPUT_LENGTH=8192
Verified signaturea9e284c -
Increase concurrency limits
Verified signature1b7add8 -
Increase concurrency: MAX_CONCURRENT_REQUESTS=8, MAX_BATCH_REQUESTS=16, MAX_BATCH_TOKENS=1024
Verified signature7947ab6 -
Upload Dockerfile with huggingface_hub
Verified signature1ccf203
-
Reduce memory: max_concurrent=1, max_batch_tokens=512
Jasai25e25b1 -
Try float16 to reduce memory usage
Jasai105f58c -
Fix: use TEI 1.9 (1.6 doesn't support Qwen3)
Jasaife6dce9 -
Deploy Qwen3-Embedding-0.6B with TEI
Jasai130b24b
-
Restore stable PyTorch CPU + 4B (backup while INT8 ONNX upload in progress)
hu3d7d8c9 -
TEI INT8 ONNX + --pooling cls --dtype float32
huecc3c86 -
TEI + INT8 ONNX Qwen3-Embedding-4B (4GB, fits cpu-basic)
hu2ce60ae
-
Stable: PyTorch CPU-only + Qwen3-Embedding-4B (proven working)
hua8af8fb -
Remove build-time model download; use ORT backend at runtime
hu73fe58d -
Add ONNX Runtime backend option + ST fallback for Qwen3-Embedding-4B
hu76a10e4 -
TEI Candle CPU + Qwen3-Embedding-4B fp16
hu624b719 -
Use CPU-only PyTorch for smaller image & faster inference
huf4aaf14 -
Switch to sentence-transformers + Qwen3-Embedding-4B (PyTorch CPU)
hu1616f63 -
Use ONNX ST Qwen3-Embedding-4B model
hu791549a -
Try TEI with Qwen3-Embedding-4B
hud2342b1 -
Switch to TEI cpu-1.9 for Qwen3-Embedding-8B
huf9a3360 -
Add Qwen3-Embedding-8B API
hue3cb2ba -
initial commit
Verified signature1027559