-
At the moment the 'Execution Layer' model view displays redundant information. It would also be useful for the user to have a clearer understanding of how each layer and base data source fits together…
-
Hi,
We have been working with the VX delegate to execute TFLite models on the NPU of the i.MX 8M Plus which is a VeriSilicon's VIPNano-SI+. Doing so, we have found that there are mismatches between…
-
### 🚀 The feature, motivation and pitch
We want to support various alignment and distillation loss functions.
Refer this PR on ORPO: #362
## Progress
### Alignment
- [x] ORPO https://gith…
-
Please add command line for send $IP and withdraw $IP staking reward.
-
Is there a plan to support the Tensor Parallel and Context Parallel? I think this would make the project more flexible and attract more users.
-
-
### Describe the issue
Lets' say I have an onnx model that takes an input 1x3x224x224. I want to change the model such that I can do batch inference. The two ways, I could do it is setting the first …
-
hi, since native comfyui already supported t5 GUFF text encoder to speed up loading Flux model [link](https://www.reddit.com/r/StableDiffusion/comments/1ewpwtp/flux1_t5_v11xxl_gguf_clip_encode_compare…
-
The current on-device certificate generation mechanism during personalization (and during ROM_EXT execution) generates DICE certificate chain in the X.509 ASN.1 DER format, which is one of the [suppor…
-
On Spring Boot 3.3.3, using Spring Boot Maven Plugin 3.3.3.
We build a Docker image using the spring-boot-maven-plugin and enabled CDS, because we measured a noticeable speedup in startup time. Unf…