From b6058e0f7e340cac295a0080fcfb913e7acdce8e Mon Sep 17 00:00:00 2001 From: "github-actions[bot]" Date: Mon, 10 Jun 2024 06:36:30 +0000 Subject: [PATCH] feat: Updated Together AI models --- .../TogetherAI/src/Predefined/AllModels.cs | 20 ++++++--- .../TogetherAI/src/TogetherAiModelIds.cs | 44 ++++++++++++++----- .../TogetherAI/src/TogetherAiModelProvider.cs | 4 +- 3 files changed, 51 insertions(+), 17 deletions(-) diff --git a/src/Providers/TogetherAI/src/Predefined/AllModels.cs b/src/Providers/TogetherAI/src/Predefined/AllModels.cs index c2149345..76b043ac 100644 --- a/src/Providers/TogetherAI/src/Predefined/AllModels.cs +++ b/src/Providers/TogetherAI/src/Predefined/AllModels.cs @@ -95,6 +95,11 @@ public class Qwen15Chat72BModel(TogetherAiProvider provider) : TogetherAiModel(p public class Qwen15Chat7BModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.Qwen15Chat7B); +/// +/// Open Router Provider Instance +public class Qwen2Instruct72BModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.Qwen2Instruct72B); + + /// /// Open Router Provider Instance public class SnowflakeArcticInstructModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.SnowflakeArcticInstruct); @@ -220,6 +225,11 @@ public class VicunaV1513BModel(TogetherAiProvider provider) : TogetherAiModel(pr public class VicunaV157BModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.VicunaV157B); +/// +/// Open Router Provider Instance +public class ReservedDbrxInstructModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.ReservedDbrxInstruct); + + /// /// Open Router Provider Instance public class Llama2Chat13BModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.Llama2Chat13B); @@ -235,11 +245,6 @@ public class Llama2Chat70BModel(TogetherAiProvider provider) : TogetherAiModel(p public class Llama2Chat7BModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.Llama2Chat7B); -/// -/// Open Router Provider Instance -public class MetaLlama370BChatModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.MetaLlama370BChat); - - /// /// Open Router Provider Instance public class MetaLlama38BChatModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.MetaLlama38BChat); @@ -330,6 +335,11 @@ public class UpstageSolarInstructV111BModel(TogetherAiProvider provider) : Toget public class _01AiYiChat34BModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds._01AiYiChat34B); +/// +/// Open Router Provider Instance +public class MetaLlama370BChatModel(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.MetaLlama370BChat); + + /// /// Open Router Provider Instance public class Llama38BChatHfInt4Model(TogetherAiProvider provider) : TogetherAiModel(provider, TogetherAiModelIds.Llama38BChatHfInt4); diff --git a/src/Providers/TogetherAI/src/TogetherAiModelIds.cs b/src/Providers/TogetherAI/src/TogetherAiModelIds.cs index 021d8182..b107009e 100644 --- a/src/Providers/TogetherAI/src/TogetherAiModelIds.cs +++ b/src/Providers/TogetherAI/src/TogetherAiModelIds.cs @@ -215,6 +215,17 @@ public enum TogetherAiModelIds /// Qwen15Chat7B, + /// + /// Name: Qwen 2 Instruct (72B)
+ /// Organization: Qwen
+ /// Context Length: 32768
+ /// Prompt Cost: $0.9/MTok
+ /// Completion Cost: $0.9/MTok
+ /// Description: Qwen2 is the new series of Qwen large language models. For Qwen2, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters, including a Mixture-of-Experts model.
+ /// HuggingFace Url: https://huggingface.co/Qwen/Qwen2-72B-Instruct + /// + Qwen2Instruct72B, + /// /// Name: Snowflake Arctic Instruct
/// Organization: Snowflake
@@ -490,6 +501,17 @@ public enum TogetherAiModelIds /// VicunaV157B, + /// + /// Name: Reserved - DBRX Instruct
+ /// Organization: Databricks
+ /// Context Length: 0
+ /// Prompt Cost: $1.2/MTok
+ /// Completion Cost: $1.2/MTok
+ /// Description: DBRX Instruct is a mixture-of-experts (MoE) large language model trained from scratch by Databricks. DBRX Instruct specializes in few-turn interactions.
+ /// HuggingFace Url: https://huggingface.co/medaltv/dbrx-instruct + /// + ReservedDbrxInstruct, + /// /// Name: LLaMA-2 Chat (13B)
/// Organization: Meta
@@ -523,17 +545,6 @@ public enum TogetherAiModelIds /// Llama2Chat7B, - /// - /// Name: Meta Llama 3 70B Chat
- /// Organization: Meta
- /// Context Length: 8192
- /// Prompt Cost: $0.9/MTok
- /// Completion Cost: $0.9/MTok
- /// Description: Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.
- /// HuggingFace Url: https://huggingface.co/meta-llama/Llama-3-70b-chat-hf - /// - MetaLlama370BChat, - /// /// Name: Meta Llama 3 8B Chat
/// Organization: Meta
@@ -732,6 +743,17 @@ public enum TogetherAiModelIds /// _01AiYiChat34B, + /// + /// Name: Meta Llama 3 70B Chat
+ /// Organization: Meta
+ /// Context Length: 8192
+ /// Prompt Cost: $0.9/MTok
+ /// Completion Cost: $0.9/MTok
+ /// Description: Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.
+ /// HuggingFace Url: https://huggingface.co/meta-llama/Llama-3-70b-chat-hf + /// + MetaLlama370BChat, + /// /// Name: Llama3 8B Chat HF INT4
/// Organization: Meta
diff --git a/src/Providers/TogetherAI/src/TogetherAiModelProvider.cs b/src/Providers/TogetherAI/src/TogetherAiModelProvider.cs index 513c9405..6eff6f15 100644 --- a/src/Providers/TogetherAI/src/TogetherAiModelProvider.cs +++ b/src/Providers/TogetherAI/src/TogetherAiModelProvider.cs @@ -28,6 +28,7 @@ public static class TogetherAiModelProvider { TogetherAiModelIds.Qwen15Chat4B, new ChatModels("Qwen/Qwen1.5-4B-Chat",32768,1.0000000000000001E-07,1.0000000000000001E-07)}, { TogetherAiModelIds.Qwen15Chat72B, new ChatModels("Qwen/Qwen1.5-72B-Chat",32768,9.000000000000001E-07,9.000000000000001E-07)}, { TogetherAiModelIds.Qwen15Chat7B, new ChatModels("Qwen/Qwen1.5-7B-Chat",32768,2.0000000000000002E-07,2.0000000000000002E-07)}, + { TogetherAiModelIds.Qwen2Instruct72B, new ChatModels("Qwen/Qwen2-72B-Instruct",32768,9.000000000000001E-07,9.000000000000001E-07)}, { TogetherAiModelIds.SnowflakeArcticInstruct, new ChatModels("Snowflake/snowflake-arctic-instruct",4096,2.4E-06,2.4E-06)}, { TogetherAiModelIds.RemmSlerpL213B, new ChatModels("Undi95/ReMM-SLERP-L2-13B",4096,3E-07,3E-07)}, { TogetherAiModelIds.ToppyM7B, new ChatModels("Undi95/Toppy-M-7B",4096,2.0000000000000002E-07,2.0000000000000002E-07)}, @@ -53,10 +54,10 @@ public static class TogetherAiModelProvider { TogetherAiModelIds.GemmaInstruct7B, new ChatModels("google/gemma-7b-it",8192,2.0000000000000002E-07,2.0000000000000002E-07)}, { TogetherAiModelIds.VicunaV1513B, new ChatModels("lmsys/vicuna-13b-v1.5",4096,3E-07,3E-07)}, { TogetherAiModelIds.VicunaV157B, new ChatModels("lmsys/vicuna-7b-v1.5",4096,2.0000000000000002E-07,2.0000000000000002E-07)}, + { TogetherAiModelIds.ReservedDbrxInstruct, new ChatModels("medaltv/dbrx-instruct",0,1.2E-06,1.2E-06)}, { TogetherAiModelIds.Llama2Chat13B, new ChatModels("meta-llama/Llama-2-13b-chat-hf",4096,2.2E-07,2.2E-07)}, { TogetherAiModelIds.Llama2Chat70B, new ChatModels("meta-llama/Llama-2-70b-chat-hf",4096,9.000000000000001E-07,9.000000000000001E-07)}, { TogetherAiModelIds.Llama2Chat7B, new ChatModels("meta-llama/Llama-2-7b-chat-hf",4096,2.0000000000000002E-07,2.0000000000000002E-07)}, - { TogetherAiModelIds.MetaLlama370BChat, new ChatModels("meta-llama/Llama-3-70b-chat-hf",8192,9.000000000000001E-07,9.000000000000001E-07)}, { TogetherAiModelIds.MetaLlama38BChat, new ChatModels("meta-llama/Llama-3-8b-chat-hf",8192,2.0000000000000002E-07,2.0000000000000002E-07)}, { TogetherAiModelIds.Wizardlm28X22b, new ChatModels("microsoft/WizardLM-2-8x22B",65536,1.2E-06,1.2E-06)}, { TogetherAiModelIds.Mistral7BInstruct, new ChatModels("mistralai/Mistral-7B-Instruct-v0.1",4096,2.0000000000000002E-07,2.0000000000000002E-07)}, @@ -75,6 +76,7 @@ public static class TogetherAiModelProvider { TogetherAiModelIds.Alpaca7B, new ChatModels("togethercomputer/alpaca-7b",2048,2.0000000000000002E-07,2.0000000000000002E-07)}, { TogetherAiModelIds.UpstageSolarInstructV111B, new ChatModels("upstage/SOLAR-10.7B-Instruct-v1.0",4096,3E-07,3E-07)}, { TogetherAiModelIds._01AiYiChat34B, new ChatModels("zero-one-ai/Yi-34B-Chat",4096,8.000000000000001E-07,8.000000000000001E-07)}, + { TogetherAiModelIds.MetaLlama370BChat, new ChatModels("meta-llama/Llama-3-70b-chat-hf",8192,9.000000000000001E-07,9.000000000000001E-07)}, { TogetherAiModelIds.Llama38BChatHfInt4, new ChatModels("togethercomputer/Llama-3-8b-chat-hf-int4",8192,2.0000000000000002E-07,2.0000000000000002E-07)}, { TogetherAiModelIds.TogethercomputerLlama38BInstructInt8, new ChatModels("togethercomputer/Llama-3-8b-chat-hf-int8",8192,2.0000000000000002E-07,2.0000000000000002E-07)}, { TogetherAiModelIds.UpstageSolarInstructV111BInt4, new ChatModels("togethercomputer/SOLAR-10.7B-Instruct-v1.0-int4",4096,3E-07,3E-07)},