-
### 🚀 The feature, motivation and pitch
This paper might be of interest: https://arxiv.org/pdf/2305.05920.pdf
This paper improves inference efficiency by determining the priority of each inference…
-
## 问题描述 / Issue Description
请简要描述您遇到的问题。 / Please briefly describe the issue you encountered.
测试 winogrande报错，mmlu正常
```text
Traceback (most recent call last):
File "/home/miniconda3/envs…
-
### 📚 The doc issue
Hello,
I'm using vllm serve to serve a open AI server like with Mistral-Small-Instruct-2409 as model. When serving I use following arguments "--enable-auto-tool-choice --tool-…
-
### Issue Type
Others
### Source
binary
### Secretflow Version
secretflow v1.8.0b0
### OS Platform and Distribution
BigCloud Enterprise Linux 8.2 (Core)
### Python version
3.10.13
### Bazel …
14ctt updated
2 months ago
-
Given now we support incremental models with splits, it will be good to also allow users to set a TTL on their data.
A common use case will be to load data for last X days for serving dashboards. Req…
-
I did a quick grep for this and it seems there is not many places where we have logic that's based on this field: https://github.com/readthedocs/readthedocs.org/pull/10042
-
https://arxiv.org/pdf/2305.05920.pdf
-
**Describe the bug**
When deploying the model, I received this error message (see screenshot below).
**To Reproduce**
Follow the instructions, deploy the model.
**Expected behavior**
The mode…
-
### Bug Description
## Sample code
```
from robyn import Robyn
from rasa.core.agent import Agent
import uuid6
app = Robyn(__file__)
agent = Agent.load("./models/20230818-091151.tar.gz")
…
-
### Configuration
```hcl
resource "databricks_model_serving" "this" {
for_each = local.serving_endpoints
provider = databricks.workspace
name = each.key
route_…