From 4caa2a5322ff4c61cb6bb141174b7fea266de5ec Mon Sep 17 00:00:00 2001 From: Sharun Date: Sat, 19 Oct 2024 01:05:45 -0400 Subject: [PATCH 1/5] feat: add more Groq models --- .../resources/models.json | 305 ++++++++++++++++++ 1 file changed, 305 insertions(+) diff --git a/extensions/inference-groq-extension/resources/models.json b/extensions/inference-groq-extension/resources/models.json index 6fce1c71b..2a779b807 100644 --- a/extensions/inference-groq-extension/resources/models.json +++ b/extensions/inference-groq-extension/resources/models.json @@ -61,6 +61,254 @@ }, "engine": "groq" }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.1-70b-versatile", + "object": "model", + "name": "Groq Llama 3.1 80b Versatile", + "version": "1.1", + "description": "Groq Llama 3.1 80b Versatile with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.1-8b-instant", + "object": "model", + "name": "Groq Llama 3.1 8b Instant", + "version": "1.1", + "description": "Groq Llama 3.1 8b with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.2-11b-text-preview", + "object": "model", + "name": "Groq Llama 3.2 11b Text Preview", + "version": "1.1", + "description": "Groq Llama 3.2 11b Text Preview with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.2-11b-vision-preview", + "object": "model", + "name": "Groq Llama 3.2 11b Vision Preview", + "version": "1.1", + "description": "Groq Llama 3.2 11b Vision Preview with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.2-1b-preview", + "object": "model", + "name": "Groq Llama 3.2 1b Preview", + "version": "1.1", + "description": "Groq Llama 3.2 1b Preview with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.2-3b-preview", + "object": "model", + "name": "Groq Llama 3.2 3b Preview", + "version": "1.1", + "description": "Groq Llama 3.2 3b Preview with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.2-90b-text-preview", + "object": "model", + "name": "Groq Llama 3.2 90b Text Preview", + "version": "1.1", + "description": "Groq Llama 3.2 90b Text Preview with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "llama-3.2-90b-vision-preview", + "object": "model", + "name": "Groq Llama 3.2 90b Vision Preview", + "version": "1.1", + "description": "Groq Llama 3.2 90b Vision Preview with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Meta", + "tags": [ + "General", + "Big Context Length" + ] + }, + "engine": "groq" + }, { "sources": [ { @@ -91,6 +339,36 @@ }, "engine": "groq" }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "gemma2-9b-it", + "object": "model", + "name": "Groq Gemma 9B Instruct", + "version": "1.2", + "description": "Groq Gemma 9b Instruct with supercharged speed!", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 8192, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "Google", + "tags": [ + "General" + ] + }, + "engine": "groq" + }, { "sources": [ { @@ -121,5 +399,32 @@ ] }, "engine": "groq" + }, + { + "sources": [ + { + "url": "https://groq.com" + } + ], + "id": "distil-whisper-large-v3-en", + "object": "model", + "name": "Groq distil-whisper-large-v3-en", + "version": "1.2", + "description": "Groq distil-whisper-large-v3-en Model from HuggingFace", + "format": "api", + "settings": {}, + "parameters": { + "max_tokens": 32768, + "temperature": 0.7, + "top_p": 0.95, + "stream": true, + "stop": [], + "frequency_penalty": 0, + "presence_penalty": 0 + }, + "metadata": { + "author": "HuggingFace" + }, + "engine": "groq" } ] \ No newline at end of file From ff46a1b009910a64ff7cab701014f5dc7adac69e Mon Sep 17 00:00:00 2001 From: Sharun Date: Sat, 19 Oct 2024 19:24:16 -0400 Subject: [PATCH 2/5] add tags to groq/distil-whisper-large-v3-en --- extensions/inference-groq-extension/resources/models.json | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/extensions/inference-groq-extension/resources/models.json b/extensions/inference-groq-extension/resources/models.json index 2a779b807..66387f4a0 100644 --- a/extensions/inference-groq-extension/resources/models.json +++ b/extensions/inference-groq-extension/resources/models.json @@ -423,7 +423,11 @@ "presence_penalty": 0 }, "metadata": { - "author": "HuggingFace" + "author": "HuggingFace", + "tags": [ + "General", + "Big Context Length" + ] }, "engine": "groq" } From 44878d6103ecf511c0a63635f8d852299c1b1231 Mon Sep 17 00:00:00 2001 From: Sharun Date: Sat, 19 Oct 2024 19:36:04 -0400 Subject: [PATCH 3/5] update max_tokens for llama-3.1-70b-versatile and fix typo --- extensions/inference-groq-extension/resources/models.json | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/extensions/inference-groq-extension/resources/models.json b/extensions/inference-groq-extension/resources/models.json index 66387f4a0..16cd24a81 100644 --- a/extensions/inference-groq-extension/resources/models.json +++ b/extensions/inference-groq-extension/resources/models.json @@ -69,13 +69,13 @@ ], "id": "llama-3.1-70b-versatile", "object": "model", - "name": "Groq Llama 3.1 80b Versatile", + "name": "Groq Llama 3.1 70b Versatile", "version": "1.1", - "description": "Groq Llama 3.1 80b Versatile with supercharged speed!", + "description": "Groq Llama 3.1 70b Versatile with supercharged speed!", "format": "api", "settings": {}, "parameters": { - "max_tokens": 8192, + "max_tokens": 8000, "temperature": 0.7, "top_p": 0.95, "stream": true, From 331e2bd35c59f32a28a35110dc3e02329c27dc18 Mon Sep 17 00:00:00 2001 From: Sharun Date: Sat, 19 Oct 2024 19:39:03 -0400 Subject: [PATCH 4/5] remove distil-whisper-large-v3-en as it does not support chat completions --- .../resources/models.json | 31 ------------------- 1 file changed, 31 deletions(-) diff --git a/extensions/inference-groq-extension/resources/models.json b/extensions/inference-groq-extension/resources/models.json index 16cd24a81..d1c8c4289 100644 --- a/extensions/inference-groq-extension/resources/models.json +++ b/extensions/inference-groq-extension/resources/models.json @@ -399,36 +399,5 @@ ] }, "engine": "groq" - }, - { - "sources": [ - { - "url": "https://groq.com" - } - ], - "id": "distil-whisper-large-v3-en", - "object": "model", - "name": "Groq distil-whisper-large-v3-en", - "version": "1.2", - "description": "Groq distil-whisper-large-v3-en Model from HuggingFace", - "format": "api", - "settings": {}, - "parameters": { - "max_tokens": 32768, - "temperature": 0.7, - "top_p": 0.95, - "stream": true, - "stop": [], - "frequency_penalty": 0, - "presence_penalty": 0 - }, - "metadata": { - "author": "HuggingFace", - "tags": [ - "General", - "Big Context Length" - ] - }, - "engine": "groq" } ] \ No newline at end of file From 72d178f3c389845092da81d4de768ff7307ae517 Mon Sep 17 00:00:00 2001 From: Sharun Date: Sat, 19 Oct 2024 19:43:33 -0400 Subject: [PATCH 5/5] update max_tokens for llama-3.1-8b-instant --- extensions/inference-groq-extension/resources/models.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/extensions/inference-groq-extension/resources/models.json b/extensions/inference-groq-extension/resources/models.json index d1c8c4289..04b60bfdd 100644 --- a/extensions/inference-groq-extension/resources/models.json +++ b/extensions/inference-groq-extension/resources/models.json @@ -106,7 +106,7 @@ "format": "api", "settings": {}, "parameters": { - "max_tokens": 8192, + "max_tokens": 8000, "temperature": 0.7, "top_p": 0.95, "stream": true,