-
Notifications
You must be signed in to change notification settings - Fork 444
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
Add Signed Process Domain (SignedPD) support and fallback in HtpBackend::Init.
status:awaiting PR reviewawaiting a PR reviewawaiting a PR reviewtype:featureFor feature requestsFor feature requestsStatus: Open.#9754 In google-ai-edge/LiteRT;- Status: Open.#9752 In google-ai-edge/LiteRT;
- Status: Open.#9708 In google-ai-edge/LiteRT;
LiteRT.js / WASM: fp16-quantized PP-OCRv5 recognizer runs 21× slower than its fp32 parent on XNNPACK
status:awaiting LiteRTerinformation is sufficient and the issue is escalatedinformation is sufficient and the issue is escalatedtype:bugBugBugStatus: Open.#9663 In google-ai-edge/LiteRT;LiteRT.js / WebGPU: Matcha-TTS flow-matching decoder output diverges from WASM (mean abs 0.34, deterministic, Apple Metal)
type:bugBugBugStatus: Open.#9662 In google-ai-edge/LiteRT;LiteRT.js / WebGPU: PP-OCRv5 recognizer deterministically misreads text vs WASM (fp32 weights too, Apple Metal)
type:bugBugBugStatus: Open.#9661 In google-ai-edge/LiteRT;pleas explain the difference between liteRT litetorch litelm and litesample
status:awaiting user responseawaiting user responseawaiting user responsetype:supportSupport IssuesSupport IssuesStatus: Open.#9631 In google-ai-edge/LiteRT;[Google Tensor SDK Beta] AOT compiler rejects INT2 (2-bit) weight quantization with INTERNAL — sub-byte FC path is 4-bit-only
type:bugBugBugtype:featureFor feature requestsFor feature requeststype:tensor sdkFor issues related to Google Tensor SDKFor issues related to Google Tensor SDKStatus: Open.#9575 In google-ai-edge/LiteRT;LiteRT.js / WebNN: Failed to execute 'conv2d' on 'MLGraphBuilder' for efficientvit Segmentation demo
status:awaiting LiteRTerinformation is sufficient and the issue is escalatedinformation is sufficient and the issue is escalatedtype:bugBugBugStatus: Open.#9548 In google-ai-edge/LiteRT;GPU delegate: explicit-dequantize int8 FULLY_CONNECTED fails kernel creation with a transposed BHWC shape mismatch (Metal / OpenCL / WebGPU)
status:awaiting LiteRTerinformation is sufficient and the issue is escalatedinformation is sufficient and the issue is escalatedtype:bugBugBugStatus: Open.#9523 In google-ai-edge/LiteRT;CompiledModel::CreateOutputBuffers creates 4 bytes size buffer for CPU accelerator
status:awaiting LiteRTerinformation is sufficient and the issue is escalatedinformation is sufficient and the issue is escalatedtype:bugBugBugStatus: Open.#9518 In google-ai-edge/LiteRT;MediaTek MT6878 NPU dispatch: SIGSEGV at first inference after "contradictory buffer requirements" warning for KV-cache index tensors
status:awaiting LiteRTerinformation is sufficient and the issue is escalatedinformation is sufficient and the issue is escalatedtype:bugBugBugStatus: Open.#9500 In google-ai-edge/LiteRT;