Skip to content

Commit

Permalink
fix: relative links
Browse files Browse the repository at this point in the history
  • Loading branch information
ishaansehgal99 committed Nov 21, 2024
1 parent 5ebc8fd commit abad1c3
Show file tree
Hide file tree
Showing 7 changed files with 37 additions and 37 deletions.
16 changes: 8 additions & 8 deletions presets/README.md
Original file line number Diff line number Diff line change
@@ -1,14 +1,14 @@
# Kaito Preset Configurations
The current supported model families with preset configurations are listed below.

| Model Family | Compatible Kaito Versions |
|-----------------------------------|---------------------------|
| [falcon](./models/falcon) | v0.0.1+|
| [llama2](./models/llama2) | v0.0.1+|
| [llama2chat](./models/llama2chat) | v0.0.1+|
| [mistral](./models/mistral) | v0.2.0+|
| [phi2](./models/phi2) | v0.2.0+|
| [phi3](./models/phi3) | v0.3.0+|
| Model Family | Compatible Kaito Versions |
|---------------------------------------------|---------------------------|
| [falcon](./workspace/models/falcon) | v0.0.1+|
| [llama2](./workspace/models/llama2) | v0.0.1+|
| [llama2chat](./workspace/models/llama2chat) | v0.0.1+|
| [mistral](./workspace/models/mistral) | v0.2.0+|
| [phi2](./workspace/models/phi2) | v0.2.0+|
| [phi3](./workspace/models/phi3) | v0.3.0+|

## Validation
Each preset model has its own hardware requirements in terms of GPU count and GPU memory defined in the respective `model.go` file. Kaito controller performs a validation check of whether the specified SKU and node count are sufficient to run the model or not. In case the provided SKU is not in the known list, the controller bypasses the validation check which means users need to ensure the model can run with the provided SKU.
Expand Down
12 changes: 6 additions & 6 deletions presets/workspace/models/falcon/README.md
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
## Supported Models
|Model name| Model source | Sample workspace|Kubernetes Workload|Distributed inference|
|----|:----:|:----:| :----: |:----: |
|falcon-7b-instruct |[tiiuae](https://huggingface.co/tiiuae/falcon-7b-instruct)|[link](../../../examples/inference/kaito_workspace_falcon_7b-instruct.yaml)|Deployment| false|
|falcon-7b |[tiiuae](https://huggingface.co/tiiuae/falcon-7b) |[link](../../../examples/inference/kaito_workspace_falcon_7b.yaml)|Deployment| false|
|falcon-40b-instruct|[tiiuae](https://huggingface.co/tiiuae/falcon-40b-instruct) |[link](../../../examples/inference/kaito_workspace_falcon_40b-instruct.yaml)|Deployment| false|
|falcon-40b |[tiiuae](https://huggingface.co/tiiuae/falcon-40b)|[link](../../../examples/inference/kaito_workspace_falcon_40b.yaml)|Deployment| false|
| Model name | Model source | Sample workspace | Kubernetes Workload | Distributed inference |
|---------------------|:-----------------------------------------------------------:|:-------------------------------------------------------------------------------:|:-------------------:|:---------------------:|
| falcon-7b-instruct | [tiiuae](https://huggingface.co/tiiuae/falcon-7b-instruct) | [link](../../../../examples/inference/kaito_workspace_falcon_7b-instruct.yaml) | Deployment | false |
| falcon-7b | [tiiuae](https://huggingface.co/tiiuae/falcon-7b) | [link](../../../../examples/inference/kaito_workspace_falcon_7b.yaml) | Deployment | false |
| falcon-40b-instruct | [tiiuae](https://huggingface.co/tiiuae/falcon-40b-instruct) | [link](../../../../examples/inference/kaito_workspace_falcon_40b-instruct.yaml) | Deployment | false |
| falcon-40b | [tiiuae](https://huggingface.co/tiiuae/falcon-40b) | [link](../../../../examples/inference/kaito_workspace_falcon_40b.yaml) | Deployment | false |

## Image Source
- **Public**: Kaito maintainers manage the lifecycle of the inference service images that contain model weights. The images are available in Microsoft Container Registry (MCR).
Expand Down
10 changes: 5 additions & 5 deletions presets/workspace/models/llama2/README.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,9 @@
## Supported Models
|Model name| Model source | Sample workspace|Kubernetes Workload|Distributed inference|
|----|:----:|:----:| :----: |:----: |
|llama2-7b |[meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md)|[link](../../../examples/inference/kaito_workspace_llama2_7b.yaml)|Deployment| false|
|llama2-13b|[meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md)|[link](../../../examples/inference/kaito_workspace_llama2_13b.yaml)|StatefulSet| true|
|llama2-70b|[meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md)|[link](../../../examples/inference/kaito_workspace_llama2_70b.yaml)|StatefulSet| true|
| Model name | Model source | Sample workspace | Kubernetes Workload | Distributed inference |
|------------|:-------------------------------------------------------------------------:|:----------------------------------------------------------------------:|:-------------------:|:---------------------:|
| llama2-7b | [meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) | [link](../../../../examples/inference/kaito_workspace_llama2_7b.yaml) | Deployment | false |
| llama2-13b | [meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) | [link](../../../../examples/inference/kaito_workspace_llama2_13b.yaml) | StatefulSet | true |
| llama2-70b | [meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) | [link](../../../../examples/inference/kaito_workspace_llama2_70b.yaml) | StatefulSet | true |

## Image Source
- **Private**: User needs to manage the lifecycle of the inference service images that contain model weights (e.g., managing image tags). The images are available in user's private container registry.
Expand Down
10 changes: 5 additions & 5 deletions presets/workspace/models/llama2chat/README.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,9 @@
## Supported Models
|Model name| Model source | Sample workspace|Kubernetes Workload|Distributed inference|
|----|:----:|:----:| :----: |:----: |
|llama2-7b-chat |[meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md)|[link](../../../examples/inference/kaito_workspace_llama2_7b-chat.yaml)|Deployment| false|
|llama2-13b-chat|[meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md)|[link](../../../examples/inference/kaito_workspace_llama2_13b-chat.yaml)|StatefulSet| true|
|llama2-70b-chat|[meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md)|[link](../../../examples/inference/kaito_workspace_llama2_70b-chat.yaml)|StatefulSet| true|
| Model name | Model source | Sample workspace | Kubernetes Workload | Distributed inference |
|-----------------|:-------------------------------------------------------------------------:|:---------------------------------------------------------------------------:|:-------------------:|:---------------------:|
| llama2-7b-chat | [meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) | [link](../../../../examples/inference/kaito_workspace_llama2_7b-chat.yaml) | Deployment | false |
| llama2-13b-chat | [meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) | [link](../../../../examples/inference/kaito_workspace_llama2_13b-chat.yaml) | StatefulSet | true |
| llama2-70b-chat | [meta](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) | [link](../../../../examples/inference/kaito_workspace_llama2_70b-chat.yaml) | StatefulSet | true |

## Image Source
- **Private**: User needs to manage the lifecycle of the inference service images that contain model weights (e.g., managing image tags). The images are available in user's private container registry.
Expand Down
8 changes: 4 additions & 4 deletions presets/workspace/models/mistral/README.md
Original file line number Diff line number Diff line change
@@ -1,8 +1,8 @@
## Supported Models
|Model name| Model source | Sample workspace|Kubernetes Workload|Distributed inference|
|----|:----:|:----:| :----: |:----: |
|mistral-7b-instruct |[mistralai](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)|[link](../../../examples/inference/kaito_workspace_mistral_7b-instruct.yaml)|Deployment| false|
|mistral-7b |[mistralai](https://huggingface.co/mistralai/Mistral-7B-v0.1) |[link](../../../examples/inference/kaito_workspace_mistral_7b.yaml)|Deployment| false|
| Model name | Model source | Sample workspace | Kubernetes Workload | Distributed inference |
|---------------------|:----------------------------------------------------------------------:|:-------------------------------------------------------------------------------:|:-------------------:|:---------------------:|
| mistral-7b-instruct | [mistralai](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) | [link](../../../../examples/inference/kaito_workspace_mistral_7b-instruct.yaml) | Deployment | false |
| mistral-7b | [mistralai](https://huggingface.co/mistralai/Mistral-7B-v0.1) | [link](../../../../examples/inference/kaito_workspace_mistral_7b.yaml) | Deployment | false |


## Image Source
Expand Down
6 changes: 3 additions & 3 deletions presets/workspace/models/phi2/README.md
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
## Supported Models
|Model name| Model source | Sample workspace|Kubernetes Workload|Distributed inference|
|----|:----:|:----:| :----: |:----: |
|phi-2 |[microsoft](https://huggingface.co/microsoft/phi-2)|[link](../../../examples/inference/kaito_workspace_phi_2.yaml)|Deployment| false|
| Model name | Model source | Sample workspace | Kubernetes Workload | Distributed inference |
|------------|:---------------------------------------------------:|:-----------------------------------------------------------------:|:-------------------:|:---------------------:|
| phi-2 | [microsoft](https://huggingface.co/microsoft/phi-2) | [link](../../../../examples/inference/kaito_workspace_phi_2.yaml) | Deployment | false |


## Image Source
Expand Down
12 changes: 6 additions & 6 deletions presets/workspace/models/phi3/README.md
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
## Supported Models
| Model name | Model source | Sample workspace|Kubernetes Workload|Distributed inference|
|--------------------------|:-----------------------------------------------------------------------:|:----:| :----: |:----: |
| phi-3-mini-4k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) |[link](../../../examples/inference/kaito_workspace_phi_3_mini_4k.yaml)|Deployment| false|
| phi-3-mini-128k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) |[link](../../../examples/inference/kaito_workspace_phi_3_mini_128k.yaml)|Deployment| false|
| phi-3-medium-4k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-medium-4k-instruct) |[link](../../../examples/inference/kaito_workspace_phi_3_medium_4k.yaml)|Deployment| false|
| phi-3-medium-128k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-medium-128k-instruct) |[link](../../../examples/inference/kaito_workspace_phi_3_medium_128k.yaml)|Deployment| false|
| Model name | Model source | Sample workspace | Kubernetes Workload | Distributed inference |
|----------------------------|:------------------------------------------------------------------------:|:-----------------------------------------------------------------------------:|:-------------------:|:---------------------:|
| phi-3-mini-4k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) | [link](../../../../examples/inference/kaito_workspace_phi_3_mini_4k.yaml) | Deployment | false |
| phi-3-mini-128k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) | [link](../../../../examples/inference/kaito_workspace_phi_3_mini_128k.yaml) | Deployment | false |
| phi-3-medium-4k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-medium-4k-instruct) | [link](../../../../examples/inference/kaito_workspace_phi_3_medium_4k.yaml) | Deployment | false |
| phi-3-medium-128k-instruct | [microsoft](https://huggingface.co/microsoft/Phi-3-medium-128k-instruct) | [link](../../../../examples/inference/kaito_workspace_phi_3_medium_128k.yaml) | Deployment | false |

## Image Source
- **Public**: Kaito maintainers manage the lifecycle of the inference service images that contain model weights. The images are available in Microsoft Container Registry (MCR).
Expand Down

0 comments on commit abad1c3

Please sign in to comment.