Skip to content

Dan v2 updates #1491

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
May 31, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 6 additions & 31 deletions pgml-cms/docs/product/cloud-database/serverless.md
Original file line number Diff line number Diff line change
Expand Up @@ -19,39 +19,14 @@ Once logged in, select "New Database" from the left menu and choose the Serverle
### Serverless Pricing
Storage is charged per GB/mo, and all requests by CPU or GPU millisecond of compute required to perform them.

#### Vector & Relational Database
| NAME | PRICING |
| :--- | ---: |
| Tables & Index Storage | $0.20 GB per month |
| Retrieval, Filtering, Ranking & other Queries | $7.50 per hour |
| Embeddings | Included w/ Queries |
| LLMs | Included w/ Queries |
| Fine Tuning | Included w/ Queries |
| Machine Learning | Included w/ Queries |

<turbo-frame id="serverless-pricing-turboframe" src="/dashboard/serverless_pricing/turboframe?style=marketing">
Loading our current pricing model...
</turbo-frame>

### Serverless Models

Serverless AI engines come with predefined models and a flexible pricing structure

#### Embedding Models
| NAME | PARAMETERS (M) | MAX INPUT TOKENS | DIMENSIONS | STRENGTHS |
| --- | --- | --- | --- | --- |
| intfloat/e5-large-v2 | 33.4 | 512 | 384 | High quality, low latency |
| mixedbread-ai/mxbai-embed-large-v1 | 334 | 512 | 1024 | High quality, higher latency |
| Alibaba-NLP/gte-base-en-v1.5 | 137 | 8192 | 768 | Supports up to 8k input tokens, low latency |
| Alibaba-NLP/gte-large-en-v1.5 | 434 | 8192 | 1024 | Supports up to 8k input tokens, higher latency |

#### Instruct Models
| NAME | TOTAL PARAMETERS (M) | ACTIVE PARAMETERS (M) | CONTEXT SIZE | STRENGTHS |
| --- | --- | --- | --- | --- |
| meta-llama/Meta-Llama-3-70B-Instruct | 70,000 | 70,000 | 8,000 | High quality |
| meta-llama/Meta-Llama-3-8B-Instruct | 8,000 | 8,000 | 8,000 | High quality, low latency |
| microsoft/Phi-3-mini-128k-instruct | 3,820 | 3,820 | 128,000 | Lowest latency |
| mistralai/Mixtral-8x7B-Instruct-v0.1 | 56,000 | 12,900 | 32,768 | MOE high quality |
| mistralai/Mistral-7B-Instruct-v0.2 | 7,000 | 7,000 | 32,768 | High quality, low latency |

#### Summarization Models
| NAME | PARAMETERS (M) | CONTEXT SIZE | STRENGTHS |
| --- | --- | --- | --- |
| google/pegasus-xsum | 568 | 512 | Efficient summarization |
<turbo-frame id="serverless-models-turboframe" src="/dashboard/serverless_models/turboframe?style=marketing">
Loading our current serverless models offered...
</turbo-frame>
93 changes: 86 additions & 7 deletions pgml-dashboard/Cargo.lock

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

8 changes: 8 additions & 0 deletions pgml-dashboard/src/components/tables/mod.rs
Original file line number Diff line number Diff line change
Expand Up @@ -4,5 +4,13 @@
// src/components/tables/large
pub mod large;

// src/components/tables/serverless_models
pub mod serverless_models;
pub use serverless_models::ServerlessModels;

// src/components/tables/serverless_pricing
pub mod serverless_pricing;
pub use serverless_pricing::ServerlessPricing;

// src/components/tables/small
pub mod small;
114 changes: 114 additions & 0 deletions pgml-dashboard/src/components/tables/serverless_models/mod.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,114 @@
use crate::components::tables::small::row::Row;
use pgml_components::component;
use pgml_components::Component;
use sailfish::TemplateOnce;

#[derive(TemplateOnce, Default)]
#[template(path = "tables/serverless_models/template.html")]
pub struct ServerlessModels {
style_type: String,
embedding_models: [Component; 4],
instruct_models: [Component; 5],
summarization_models: [Component; 1],
}

impl ServerlessModels {
pub fn new() -> ServerlessModels {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This feels like it belongs in our main app, not in the open source dashboard.

Copy link
Contributor Author

@chillenberger chillenberger May 31, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I understand. That requires us to fix our architecture to be able to inject data into the docs which may not be available in the open source repo.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ServerlessModels {
style_type: "product".to_string(),
embedding_models: [
Component::from(Row::new(&[
"intfloat/e5-small-v2".into(),
"33.4".into(),
"512".into(),
"384".into(),
"Good quality, low latency".into(),
])),
Component::from(Row::new(&[
"mixedbread-ai/mxbai-embed-large-v1".into(),
"335".into(),
"512".into(),
"1024".into(),
"High quality, higher latency".into(),
])),
Component::from(Row::new(&[
"Alibaba-NLP/gte-base-en-v1.5".into(),
"137".into(),
"8192".into(),
"768".into(),
"Supports up to 8,000 input tokens".into(),
])),
Component::from(Row::new(&[
"Alibaba-NLP/gte-large-en-v1.5".into(),
"434".into(),
"8192".into(),
"1024".into(),
"Highest quality, 8,000 input tokens".into(),
])),
],
instruct_models: [
Component::from(Row::new(&[
"meta-llama/Meta-Llama-3-70B-Instruct".into(),
"70,000".into(),
"70,000".into(),
"8,000".into(),
"Highest quality".into(),
])),
Component::from(Row::new(&[
"meta-llama/Meta-Llama-3-8B-Instruct".into(),
"8,000".into(),
"8,000".into(),
"8,000".into(),
"High quality, low latency".into(),
])),
Component::from(Row::new(&[
"microsoft/Phi-3-mini-128k-instruct".into(),
"3,820".into(),
"3,820".into(),
"128,000".into(),
"Lowest latency".into(),
])),
Component::from(Row::new(&[
"mistralai/Mixtral-8x7B-Instruct-v0.1".into(),
"56,000".into(),
"12,900".into(),
"32,768".into(),
"MOE high quality".into(),
])),
Component::from(Row::new(&[
"mistralai/Mistral-7B-Instruct-v0.2".into(),
"7,000".into(),
"7,000".into(),
"32,768".into(),
"High quality, low latency".into(),
])),
],
summarization_models: [Component::from(Row::new(&[
"google/pegasus-xsum".into(),
"568".into(),
"512".into(),
"8,000".into(),
]))],
}
}

pub fn set_style_type(mut self, style_type: &str) -> Self {
self.style_type = style_type.to_string();
self
}
}

#[derive(TemplateOnce, Default)]
#[template(path = "tables/serverless_models/turbotemplate.html")]
pub struct ServerlessModelsTurbo {
comp: Component,
}

impl ServerlessModelsTurbo {
pub fn new(comp: Component) -> ServerlessModelsTurbo {
ServerlessModelsTurbo { comp }
}
}

component!(ServerlessModels);
component!(ServerlessModelsTurbo);
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
div[data-controller="tables-serverless-models"] {
table.table.table-sm thead th, table.table-sm thead th {
vertical-align: top;
padding-top: 8px;
padding-bottom: 8px;
}
}
Loading
pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy