Skip to content
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.

Sign up for GitHub

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

Added display name for models #1657

Merged
merged 2 commits into from
Nov 5, 2024
Merged

Added display name for models #1657

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
abd3e38
Added display name for models
kathirsvn Nov 5, 2024
98d069e
Added display name for models
kathirsvn Nov 5, 2024
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
3 changes: 3 additions & 0 deletions ...va/io/stargate/sgv2/jsonapi/service/embedding/configuration/EmbeddingProvidersConfig.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,6 +105,9 @@ interface ModelConfig {
@JsonProperty
String name();

@JsonProperty
String displayName();

/**
* vectorDimension is not null if the model supports a single dimension value. It will be null
* if the model supports different dimensions. A parameter called vectorDimension is included.
Expand Down
2 changes: 2 additions & 0 deletions ...o/stargate/sgv2/jsonapi/service/embedding/configuration/EmbeddingProvidersConfigImpl.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@ public record TokenConfigImpl(String accepted, String forwarded) implements Toke

public record ModelConfigImpl(
String name,
String displayName,
Optional<Integer> vectorDimension,
List<ParameterConfig> parameters,
Map<String, String> properties,
Expand All @@ -38,6 +39,7 @@ public ModelConfigImpl(
List<ParameterConfig> modelParameterList) {
this(
grpcModelConfig.getName(),
grpcModelConfig.getDisplayName(),
grpcModelConfig.hasVectorDimension()
? Optional.of(grpcModelConfig.getVectorDimension())
: Optional.empty(),
Expand Down
1 change: 1 addition & 0 deletions src/main/proto/embedding_gateway.proto
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -167,6 +167,7 @@ message GetSupportedProvidersResponse {
map<string, string> properties = 4;
optional bool deprecated = 5;
optional string serviceUrl = 6;
string display_name = 7;
kathirsvn marked this conversation as resolved.
Show resolved Hide resolved
}
}

Expand Down
33 changes: 33 additions & 0 deletions src/main/resources/embedding-providers-config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,7 @@ stargate:
max-batch-size: 2048
models:
- name: text-embedding-3-small
display-name: "text-embedding-3-small"
parameters:
- name: vectorDimension
type: number
Expand All @@ -47,6 +48,7 @@ stargate:
numeric-range: [2, 1536]
help: "Vector dimension to use in the database and when calling OpenAI."
- name: text-embedding-3-large
display-name: "text-embedding-3-large"
parameters:
- name: vectorDimension
type: number
Expand All @@ -56,6 +58,7 @@ stargate:
numeric-range: [256, 3072]
help: "Vector dimension to use in the database and when calling OpenAI."
- name: text-embedding-ada-002
display-name: "text-embedding-ada-002"
vector-dimension: 1536
azureOpenAI:
# see https://learn.microsoft.com/en-us/azure/ai-services/openai/reference
Expand Down Expand Up @@ -92,6 +95,7 @@ stargate:
max-batch-size: 2048
models:
- name: text-embedding-3-small
display-name: "text-embedding-3-small"
parameters:
- name: vectorDimension
type: number
Expand All @@ -101,6 +105,7 @@ stargate:
numeric-range: [2, 1536]
help: "Vector dimension to use in the database and when calling Azure OpenAI."
- name: text-embedding-3-large
display-name: "text-embedding-3-large"
parameters:
- name: vectorDimension
type: number
Expand All @@ -113,6 +118,7 @@ stargate:
numeric-range: [256, 3072]
help: "Vector dimension to use in the database and when calling Azure OpenAI."
- name: text-embedding-ada-002
display-name: "text-embedding-ada-002"
vector-dimension: 1536
bedrock:
display-name: Amazon Bedrock
Expand Down Expand Up @@ -143,8 +149,10 @@ stargate:
max-batch-size: 1
models:
- name: amazon.titan-embed-text-v1
display-name: "amazon.titan-embed-text-v1"
vector-dimension: 1536
- name: amazon.titan-embed-text-v2:0
display-name: "amazon.titan-embed-text-v2:0"
parameters:
- name: vectorDimension
type: number
Expand Down Expand Up @@ -175,16 +183,22 @@ stargate:
max-batch-size: 32
models:
- name: sentence-transformers/all-MiniLM-L6-v2
display-name: "sentence-transformers/all-MiniLM-L6-v2"
vector-dimension: 384
- name: intfloat/multilingual-e5-large
display-name: "intfloat/multilingual-e5-large"
vector-dimension: 1024
- name: intfloat/multilingual-e5-large-instruct
display-name: "intfloat/multilingual-e5-large-instruct"
vector-dimension: 1024
- name: BAAI/bge-small-en-v1.5
display-name: "BAAI/bge-small-en-v1.5"
vector-dimension: 384
- name: BAAI/bge-base-en-v1.5
display-name: "BAAI/bge-base-en-v1.5"
vector-dimension: 768
- name: BAAI/bge-large-en-v1.5
display-name: "BAAI/bge-large-en-v1.5"
vector-dimension: 1024
huggingfaceDedicated:
# see https://huggingface.co/docs/inference-endpoints/en/supported_tasks#sentence-embeddings
Expand All @@ -208,6 +222,7 @@ stargate:
max-batch-size: 32
models:
- name: endpoint-defined-model
display-name: "endpoint-defined-model"
parameters:
- name: vectorDimension
type: number
Expand Down Expand Up @@ -265,6 +280,7 @@ stargate:
max-batch-size: 32
models:
- name: textembedding-gecko@003
display-name: "textembedding-gecko@003"
vector-dimension: 768
parameters:
- name: "autoTruncate"
Expand Down Expand Up @@ -295,8 +311,10 @@ stargate:
max-batch-size: 32
models:
- name: embed-english-v3.0
display-name: "embed-english-v3.0"
vector-dimension: 1024
- name: embed-english-v2.0
display-name: "embed-english-v2.0"
vector-dimension: 4096
nvidia:
display-name: Nvidia
Expand All @@ -309,6 +327,7 @@ stargate:
max-batch-size: 8
models:
- name: NV-Embed-QA
display-name: "NV-Embed-QA"
vector-dimension: 1024
properties:
max-tokens: 512
Expand Down Expand Up @@ -336,14 +355,19 @@ stargate:
max-batch-size: 32
models:
- name: jina-embeddings-v2-base-en
display-name: "jina-embeddings-v2-base-en"
vector-dimension: 768
- name: jina-embeddings-v2-base-de
display-name: "jina-embeddings-v2-base-de"
vector-dimension: 768
- name: jina-embeddings-v2-base-es
display-name: "jina-embeddings-v2-base-es"
vector-dimension: 768
- name: jina-embeddings-v2-base-code
display-name: "jina-embeddings-v2-base-code"
vector-dimension: 768
- name: jina-embeddings-v2-base-zh
display-name: "jina-embeddings-v2-base-zh"
vector-dimension: 768
voyageAI:
# see https://docs.voyageai.com/reference/embeddings-api
Expand Down Expand Up @@ -377,18 +401,25 @@ stargate:
max-batch-size: 32
models:
- name: voyage-large-2-instruct
display-name: "voyage-large-2-instruct"
vector-dimension: 1024
- name: voyage-law-2
display-name: "voyage-law-2"
vector-dimension: 1024
- name: voyage-code-2
display-name: "voyage-code-2"
vector-dimension: 1536
- name: voyage-large-2
display-name: "voyage-large-2"
vector-dimension: 1536
- name: voyage-2
display-name: "voyage-2"
vector-dimension: 1024
- name: voyage-finance-2
display-name: "voyage-finance-2"
vector-dimension: 1024
- name: voyage-multilingual-2
display-name: "voyage-multilingual-2"
vector-dimension: 1024
mistral:
# see https://docs.mistral.ai/api/#operation/createEmbedding
Expand All @@ -413,6 +444,7 @@ stargate:
max-batch-size: 32
models:
- name: mistral-embed
display-name: "mistral-embed"
vector-dimension: 1024

# NOTE: UpstageAI has one model for storing and a diff one for reading: this is different
Expand Down Expand Up @@ -444,4 +476,5 @@ stargate:
# NOTE: this is where weirdness exists; model name is prefix on which
# either "-query" or "-passage" is appended to get the actual model name
- name: solar-embedding-1-large
display-name: "solar-embedding-1-large"
vector-dimension: 4096