Nicholai/jan
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(spec): model spec update

Browse Source
This commit is contained in:
hiro 2023-11-18 17:04:51 +07:00
parent 361b909a55
commit df883a7cb8
1 changed files with 166 additions and 75 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

241
docs/docs/docs/specs/models.md
View File

@ -46,7 +46,7 @@ _Users can override run settings at runtime_
| `object` | enum: `model`, `assistant`, `thread`, `message` | Type of the Jan Object. Always `model` | Defaults to "model" | | `object` | enum: `model`, `assistant`, `thread`, `message` | Type of the Jan Object. Always `model` | Defaults to "model" |
| `name` | string | A vanity name | Defaults to filename | | `name` | string | A vanity name | Defaults to filename |
| `description` | string | A vanity description of the model | Defaults to "" | | `description` | string | A vanity description of the model | Defaults to "" |
| `state` | enum[`running` , `stopped`, `not-downloaded` , `downloading`] | Needs more thought | Defaults to `not-downloaded` | | `state` | enum[`to_download` , `downloading`, `ready` , `running`] | Needs more thought | Defaults to `to_download` |
| `parameters` | map | Defines default model run parameters used by any assistant. | Defaults to `{}` | | `parameters` | map | Defines default model run parameters used by any assistant. | Defaults to `{}` |
| `metadata` | map | Stores additional structured information about the model. | Defaults to `{}` | | `metadata` | map | Stores additional structured information about the model. | Defaults to `{}` |
| `metadata.engine` | enum: `llamacpp`, `api`, `tensorrt` | The model backend used to run model. | Defaults to "llamacpp" | | `metadata.engine` | enum: `llamacpp`, `api`, `tensorrt` | The model backend used to run model. | Defaults to "llamacpp" |
@ -83,10 +83,10 @@ Additionally, Jan supports importing popular formats. For example, if you provid
Supported URL formats with custom importers: Supported URL formats with custom importers:
- `huggingface/thebloke`: `TODO: URL here` - `huggingface/thebloke`: [Link](https://huggingface.co/TheBloke/Llama-2-7B-GGUF)
- `janhq`: `TODO: put URL here` - `janhq`: `TODO: put URL here`
- `azure_openai`: `TODO: put URL here` - `azure_openai`: `https://docs-test-001.openai.azure.com/openai.azure.com/docs-test-001/gpt4-turbo`
- `openai`: `TODO: put URL here` - `openai`: `api.openai.com`
### Generic Example ### Generic Example
@ -98,52 +98,66 @@ Supported URL formats with custom importers:
// Note: Default fields omitted for brevity // Note: Default fields omitted for brevity
"source_url": "https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/blob/main/zephyr-7b-beta.Q4_K_M.gguf", "source_url": "https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/blob/main/zephyr-7b-beta.Q4_K_M.gguf",
"parameters": { "parameters": {
"ctx_len": 2048, "init": {
"ngl": 100, "ctx_len": "2048",
"embedding": true, "ngl": "100",
"n_parallel": 4, "embedding": "true",
"n_parallel": "4",
"pre_prompt": "A chat between a curious user and an artificial intelligence", "pre_prompt": "A chat between a curious user and an artificial intelligence",
"user_prompt": "USER: ", "user_prompt": "USER: ",
"ai_prompt": "ASSISTANT: " "ai_prompt": "ASSISTANT: "
},
"runtime": {
"temperature": "0.7", "temperature": "0.7",
"token_limit": "2048", "token_limit": "2048",
"top_k": "..", "top_k": "0",
"top_p": "..", "top_p": "1",
"stream": "true"
}
}, },
"metadata": { "metadata": {
"quantization": "..", "engine": "llamacpp",
"size": "..", "quantization": "Q3_K_L",
"size": "7B",
} }
``` ```
### Example: multiple binaries ### Example: multiple binaries
- Model has multiple binaries - Model has multiple binaries `model-llava-1.5-ggml.json`
- See [source](https://huggingface.co/mys/ggml_llava-v1.5-13b) - See [source](https://huggingface.co/mys/ggml_llava-v1.5-13b)
```json ```json
"source_url": "https://huggingface.co/mys/ggml_llava-v1.5-13b" "source_url": "https://huggingface.co/mys/ggml_llava-v1.5-13b",
"parameters": {"init": {}, "runtime": {}}
"metadata": { "metadata": {
"binaries": "..", // TODO: what should this property be "mmproj_binary": "https://huggingface.co/mys/ggml_llava-v1.5-13b/blob/main/mmproj-model-f16.gguf",
"ggml_binary": "https://huggingface.co/mys/ggml_llava-v1.5-13b/blob/main/ggml-model-q5_k.gguf",
"engine": "llamacpp",
"quantization": "Q5_K"
} }
``` ```
### Example: Azure API ### Example: Azure API
- Using a remote API to access model - Using a remote API to access model `model-azure-openai-gpt4-turbo.json`
- See [source](https://learn.microsoft.com/en-us/azure/ai-services/openai/quickstart?tabs=command-line%2Cpython&pivots=rest-api) - See [source](https://learn.microsoft.com/en-us/azure/ai-services/openai/quickstart?tabs=command-line%2Cpython&pivots=rest-api)
```json ```json
"source_url": "https://docs-test-001.openai.azure.com/openai.azure.com/docs-test-001/gpt4-turbo", "source_url": "https://docs-test-001.openai.azure.com/openai.azure.com/docs-test-001/gpt4-turbo",
"parameters": { "parameters": {
"init" {
"API-KEY": "", "API-KEY": "",
"DEPLOYMENT-NAME": "", "DEPLOYMENT-NAME": "",
"api-version": "2023-05-15", "api-version": "2023-05-15"
},
"runtime": {
"temperature": "0.7", "temperature": "0.7",
"max_tokens": "2048", "max_tokens": "2048",
"presence_penalty": "0", "presence_penalty": "0",
"top_p": "1", "top_p": "1",
"stream": "true" "stream": "true"
}
} }
"metadata": { "metadata": {
"engine": "api", "engine": "api",
@ -155,7 +169,7 @@ Supported URL formats with custom importers:
- Everything needed to represent a `model` is packaged into an `Model folder`. - Everything needed to represent a `model` is packaged into an `Model folder`.
- The `folder` is standalone and can be easily zipped, imported, and exported, e.g. to Github. - The `folder` is standalone and can be easily zipped, imported, and exported, e.g. to Github.
- The `folder` always contains at least one `Model Object`, declared in a `json` format. - The `folder` always contains at least one `Model Object`, declared in a `json` format.
- The `folder` and `file` do not have to share the same name - The `folder` and `file` do not have to share the same name
- The model `id` is made up of `folder_name/filename` and is thus always unique. - The model `id` is made up of `folder_name/filename` and is thus always unique.
```sh ```sh
@ -170,11 +184,9 @@ Supported URL formats with custom importers:
``` ```
### Default ./model folder ### Default ./model folder
- Jan ships with a default model folders containing recommended models - Jan ships with a default model folders containing recommended models
- Only the Model Object `json` files are included - Only the Model Object `json` files are included
- Users must later explicitly download the model binaries - Users must later explicitly download the model binaries
```sh ```sh
models/ models/
mistral-7b/ mistral-7b/
@ -182,7 +194,6 @@ models/
hermes-7b/ hermes-7b/
hermes-7b.json hermes-7b.json
``` ```
### Multiple quantizations ### Multiple quantizations
- Each quantization has its own `Jan Model Object` file - Each quantization has its own `Jan Model Object` file
@ -193,7 +204,6 @@ llama2-7b-gguf/
llama2-7b-gguf-Q3_K_L.json llama2-7b-gguf-Q3_K_L.json
.bin .bin
``` ```
### Multiple model partitions ### Multiple model partitions
- A Model that is partitioned into several binaries use just 1 file - A Model that is partitioned into several binaries use just 1 file
@ -204,8 +214,7 @@ llava-ggml/
.proj .proj
ggml ggml
``` ```
### Your locally fine-tuned model
### ?? whats this example for?
- ?? - ??
@ -214,67 +223,149 @@ llama-70b-finetune/
llama-70b-finetune-q5.json llama-70b-finetune-q5.json
.bin .bin
``` ```
## Jan API ## Jan API
### Model API Object ### Model API Object
- The `Jan Model Object` maps into the `OpenAI Model Object`. - The `Jan Model Object` maps into the `OpenAI Model Object`.
- Properties marked with `*` are compatible with the [OpenAI `model` object](https://platform.openai.com/docs/api-reference/models) - Properties marked with `*` are compatible with the [OpenAI `model` object](https://platform.openai.com/docs/api-reference/models)
- Note: The `Jan Model Object` has additional properties when retrieved via its API endpoint. - Note: The `Jan Model Object` has additional properties when retrieved via its API endpoint.
- https://platform.openai.com/docs/api-reference/models/object > OpenAI Equivalent: https://platform.openai.com/docs/api-reference/models/object
| Property | Type | Public Description | Jan Model Object (`m`) Property | ### Model lifecycle
| ------------- | -------------- | ----------------------------------------------------------- | -------------------------------------------- | Model has 4 states (enum)
| `id`\* | string | Model uuid; also the file location under `/models` | `folder/filename` | - `to_download`
| `object`\* | string | Always "model" | `m.object` | - `downloading`
| `created`\* | integer | Timestamp when model was created. | `m.json` creation time | - `ready`
| `owned_by`\* | string | The organization that owns the model. | grep author from `m.source_url` OR $(whoami) | - `running`
| `name` | string or null | A display name | `m.name` or filename |
| `description` | string | A vanity description of the model | `m.description` |
| `state` | enum | | |
| `parameters` | map | Defines default model run parameters used by any assistant. | |
| `metadata` | map | Stores additional structured information about the model. | |
### List models
- https://platform.openai.com/docs/api-reference/models/list
TODO: @hiro
### Get Model ### Get Model
> OpenAI Equivalent: https://platform.openai.com/docs/api-reference/models/retrieve
- https://platform.openai.com/docs/api-reference/models/retrieve - Example request
```shell
TODO: @hiro curl {JAN_URL}/v1/models/{model_id}
```
- Example response
```json
{
"id": "model-zephyr-7B",
"object": "model",
"created_at": 1686935002,
"owned_by": "thebloke",
"state": "running",
"source_url": "https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/blob/main/zephyr-7b-beta.Q4_K_M.gguf",
"parameters": {
"ctx_len": 2048,
"ngl": 100,
"embedding": true,
"n_parallel": 4,
"pre_prompt": "A chat between a curious user and an artificial intelligence",
"user_prompt": "USER: ",
"ai_prompt": "ASSISTANT: ",
"temperature": "0.7",
"token_limit": "2048",
"top_k": "0",
"top_p": "1",
},
"metadata": {
"engine": "llamacpp",
"quantization": "Q3_K_L",
"size": "7B",
}
}
```
### List models
Lists the currently available models, and provides basic information about each one such as the owner and availability.
> OpenAI Equivalent: https://platform.openai.com/docs/api-reference/models/list
- Example request
```shell=
curl {JAN_URL}/v1/models
```
- Example response
```json
{
"object": "list",
"data": [
{
"id": "model-zephyr-7B",
"object": "model",
"created_at": 1686935002,
"owned_by": "thebloke",
"state": "running"
},
{
"id": "ft-llama-70b-gguf",
"object": "model",
"created_at": 1686935002,
"owned_by": "you",
"state": "stopped"
},
{
"id": "model-azure-openai-gpt4-turbo",
"object": "model",
"created_at": 1686935002,
"owned_by": "azure_openai",
"state": "running"
},
],
"object": "list"
}
```
### Delete Model ### Delete Model
> OpenAI Equivalent: https://platform.openai.com/docs/api-reference/models/delete
- https://platform.openai.com/docs/api-reference/models/delete `- Example request
```shell
TODO: @hiro curl -X DELETE {JAN_URL}/v1/models/{model_id}
```
### Get Model State - Example response
```json
> Jan-only endpoint {
> TODO: @hiro "id": "model-zephyr-7B",
"object": "model",
### Get Model Metadata "deleted": true,
"state": "to_download"
> Jan-only endpoint }
> TODO: @hiro ```
### Download Model
> Jan-only endpoint
> TODO: @hiro
### Start Model ### Start Model
> Jan-only endpoint > Jan-only endpoint
> TODO: @hiro The request to start `model` by changing model state from `ready` to `running`
- Example request
```shell
curl -X PUT {JAN_URL}/v1/models{model_id}/start
```
- Example response
```json
{
"id": "model-zephyr-7B",
"object": "model",
"state": "running"
}
```
### Stop Model ### Stop Model
> Jan-only endpoint > Jan-only endpoint
> TODO: @hiro The request to start `model` by changing model state from `running` to `ready`
- Example request
```shell
curl -X PUT {JAN_URL}/v1/models/{model_id}/stop
```
- Example response
```json
{
"id": "model-zephyr-7B",
"object": "model",
"state": "ready"
}
```
### Download Model
> Jan-only endpoint
The request to download `model` by changing model state from `to_download` to `downloading` then `ready`once it's done.
- Example request
```shell
curl -X POST {JAN_URL}/v1/models/
```
- Example response
```json
{
"id": "model-zephyr-7B",
"object": "model",
"state": "downloading"
}
```