From 23ed15d19f138fd0bf555f17303fdafb1d74c3f7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E5=B0=8F=E7=BE=BD?= Date: Tue, 6 Aug 2024 02:16:41 +0000 Subject: [PATCH] feat:nvidia add nemotron4-340b and microsoft/phi-3 (#6973) --- .../model_providers/nvidia/llm/_position.yaml | 3 ++ .../model_providers/nvidia/llm/llm.py | 6 ++-- .../nvidia/llm/nemotron-4-340b-instruct.yaml | 36 +++++++++++++++++++ .../llm/phi-3-medium-128k-instruct.yaml | 36 +++++++++++++++++++ .../nvidia/llm/phi-3-mini-128k-instruct.yaml | 36 +++++++++++++++++++ 5 files changed, 115 insertions(+), 2 deletions(-) create mode 100644 api/core/model_runtime/model_providers/nvidia/llm/nemotron-4-340b-instruct.yaml create mode 100644 api/core/model_runtime/model_providers/nvidia/llm/phi-3-medium-128k-instruct.yaml create mode 100644 api/core/model_runtime/model_providers/nvidia/llm/phi-3-mini-128k-instruct.yaml diff --git a/api/core/model_runtime/model_providers/nvidia/llm/_position.yaml b/api/core/model_runtime/model_providers/nvidia/llm/_position.yaml index 6cc197b70b7031..ad01d430d61c79 100644 --- a/api/core/model_runtime/model_providers/nvidia/llm/_position.yaml +++ b/api/core/model_runtime/model_providers/nvidia/llm/_position.yaml @@ -10,5 +10,8 @@ - mistralai/mistral-large - mistralai/mixtral-8x7b-instruct-v0.1 - mistralai/mixtral-8x22b-instruct-v0.1 +- nvidia/nemotron-4-340b-instruct +- microsoft/phi-3-medium-128k-instruct +- microsoft/phi-3-mini-128k-instruct - fuyu-8b - snowflake/arctic diff --git a/api/core/model_runtime/model_providers/nvidia/llm/llm.py b/api/core/model_runtime/model_providers/nvidia/llm/llm.py index 494b7374f5ae28..bc42eaca658bac 100644 --- a/api/core/model_runtime/model_providers/nvidia/llm/llm.py +++ b/api/core/model_runtime/model_providers/nvidia/llm/llm.py @@ -34,8 +34,10 @@ class NVIDIALargeLanguageModel(OAIAPICompatLargeLanguageModel): 'meta/llama-3.1-8b-instruct': '', 'meta/llama-3.1-70b-instruct': '', 'meta/llama-3.1-405b-instruct': '', - 'google/recurrentgemma-2b': '' - + 'google/recurrentgemma-2b': '', + 'nvidia/nemotron-4-340b-instruct': '', + 'microsoft/phi-3-medium-128k-instruct':'', + 'microsoft/phi-3-mini-128k-instruct':'' } def _invoke(self, model: str, credentials: dict, diff --git a/api/core/model_runtime/model_providers/nvidia/llm/nemotron-4-340b-instruct.yaml b/api/core/model_runtime/model_providers/nvidia/llm/nemotron-4-340b-instruct.yaml new file mode 100644 index 00000000000000..e5537cd2fd9dc8 --- /dev/null +++ b/api/core/model_runtime/model_providers/nvidia/llm/nemotron-4-340b-instruct.yaml @@ -0,0 +1,36 @@ +model: nvidia/nemotron-4-340b-instruct +label: + zh_Hans: nvidia/nemotron-4-340b-instruct + en_US: nvidia/nemotron-4-340b-instruct +model_type: llm +features: + - agent-thought +model_properties: + mode: chat + context_size: 131072 +parameter_rules: + - name: temperature + use_template: temperature + min: 0 + max: 1 + default: 0.5 + - name: top_p + use_template: top_p + min: 0 + max: 1 + default: 1 + - name: max_tokens + use_template: max_tokens + min: 1 + max: 4096 + default: 1024 + - name: frequency_penalty + use_template: frequency_penalty + min: -2 + max: 2 + default: 0 + - name: presence_penalty + use_template: presence_penalty + min: -2 + max: 2 + default: 0 diff --git a/api/core/model_runtime/model_providers/nvidia/llm/phi-3-medium-128k-instruct.yaml b/api/core/model_runtime/model_providers/nvidia/llm/phi-3-medium-128k-instruct.yaml new file mode 100644 index 00000000000000..0c5538d1350613 --- /dev/null +++ b/api/core/model_runtime/model_providers/nvidia/llm/phi-3-medium-128k-instruct.yaml @@ -0,0 +1,36 @@ +model: microsoft/phi-3-medium-128k-instruct +label: + zh_Hans: microsoft/phi-3-medium-128k-instruct + en_US: microsoft/phi-3-medium-128k-instruct +model_type: llm +features: + - agent-thought +model_properties: + mode: chat + context_size: 131072 +parameter_rules: + - name: temperature + use_template: temperature + min: 0 + max: 1 + default: 0.5 + - name: top_p + use_template: top_p + min: 0 + max: 1 + default: 1 + - name: max_tokens + use_template: max_tokens + min: 1 + max: 4096 + default: 1024 + - name: frequency_penalty + use_template: frequency_penalty + min: -2 + max: 2 + default: 0 + - name: presence_penalty + use_template: presence_penalty + min: -2 + max: 2 + default: 0 diff --git a/api/core/model_runtime/model_providers/nvidia/llm/phi-3-mini-128k-instruct.yaml b/api/core/model_runtime/model_providers/nvidia/llm/phi-3-mini-128k-instruct.yaml new file mode 100644 index 00000000000000..1eb1c51d01157c --- /dev/null +++ b/api/core/model_runtime/model_providers/nvidia/llm/phi-3-mini-128k-instruct.yaml @@ -0,0 +1,36 @@ +model: microsoft/phi-3-mini-128k-instruct +label: + zh_Hans: microsoft/phi-3-mini-128k-instruct + en_US: microsoft/phi-3-mini-128k-instruct +model_type: llm +features: + - agent-thought +model_properties: + mode: chat + context_size: 131072 +parameter_rules: + - name: temperature + use_template: temperature + min: 0 + max: 1 + default: 0.5 + - name: top_p + use_template: top_p + min: 0 + max: 1 + default: 1 + - name: max_tokens + use_template: max_tokens + min: 1 + max: 4096 + default: 1024 + - name: frequency_penalty + use_template: frequency_penalty + min: -2 + max: 2 + default: 0 + - name: presence_penalty + use_template: presence_penalty + min: -2 + max: 2 + default: 0