From 10ffb53da83a9a0ae999e41b9eb260c4578e0ea7 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 18:49:50 +0100 Subject: [PATCH 01/14] Update models.json Add support of Orca-2 Add support Open LLaMa Add Neural-Chat --- api/src/serge/data/models.json | 81 ++++++++++++++++++++++++++++++++++ 1 file changed, 81 insertions(+) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 67102689341..6c938618dca 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -1,4 +1,85 @@ [ + { + "name": "Neural-Chat", + "models": [ + { + "name": "Neural-Chat-7B-v3-1", + "repo": "TheBloke/neural-chat-7B-v3-1-GGUFF", + "files": [ + { + "name": "q4_K_M", + "filename": " neural-chat-7b-v3-1.Q4_K_M.gguf", + "disk_space": 4370000000.0 + } + ] + } + ] + }, + { + "name": "Open_LLaMA", + "models": [ + { + "name": "Open_LLaMA-3B-v2", + "repo": "maddes8cht/openlm-research-open_llama_3b_v2-gguf", + "files": [ + { + "name": "q4_K_M", + "filename": "openlm-research-open_llama_3b_v2-Q4_K_M.gguff", + "disk_space": 2580000000.0 + } + ] + }, + { + "name": "Open_LLaMA-7B-v2", + "repo": "maddes8cht/openlm-research-open_llama_7b_v2-gguf", + "files": [ + { + "name": "q4_K_M", + "filename": "openlm-research-open_llama_7b_v2-Q4_K_M.gguf", + "disk_space": 4080000000.0 + } + ] + }, + { + "name": "Open_LLaMA-13B-V2", + "repo": "maddes8cht/openlm-research-open_llama_13b-gguf", + "files": [ + { + "name": "q4_K_M", + "filename": "openlm-research-open_llama_13b-Q4_K_M.gguf", + "disk_space": 7870000000.0 + } + ] + } + ] + }, + { + "name": "Orca-2", + "models": [ + { + "name": "Orca-2-7B", + "repo": "TheBloke/Orca-2-7B-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": " orca-2-7b.Q4_K_M.gguf", + "disk_space": 4080000000.0 + } + ] + }, + { + "name": "Orca-2-13B", + "repo": "TheBloke/Orca-2-13B-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": " orca-2-13b.Q4_K_M.gguf", + "disk_space": 7870000000.0 + } + ] + } + ] + }, { "name": "Meditron", "models": [ From 50398f8a6478316de17a67ebc621a10cf2068442 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 18:51:55 +0100 Subject: [PATCH 02/14] Update README.md --- README.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/README.md b/README.md index e0c180ee45e..245d4e26c1c 100644 --- a/README.md +++ b/README.md @@ -62,6 +62,9 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **LLaMA** | 7B, 13B, 70B | | **Meditron** | 7B, 70B | | **Mistral** | 7B, 7B-Instruct, 7B-OpenOrca | +| **Neural-Chat ** | 7B-v3.1 | +| **OpenLLaMA** | 3B-v2, 7B-v2, 13B-v2 | +| **Orca-2** | 7B, 13B | | **Vicuna** | 7B-v1.5, 13B-v1.5 | | **Zephyr** | 7B-Alpha, 7B-Beta | From f5ec8e32fbd71c1a929e42cf6ac885dca298104e Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 19:26:03 +0100 Subject: [PATCH 03/14] Update README.md add Falcon 180B and Falcon 180B-Chat --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 245d4e26c1c..3443bb82c15 100644 --- a/README.md +++ b/README.md @@ -58,7 +58,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | Category | Models | |:-------------:|:-------| | **CodeLLaMA** | 7B, 13B | -| **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct | +| **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct, 180B, 180b-Chat | | **LLaMA** | 7B, 13B, 70B | | **Meditron** | 7B, 70B | | **Mistral** | 7B, 7B-Instruct, 7B-OpenOrca | From bb110792e6df95e61b106f795ce99a1f98078d0e Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 22:31:25 +0100 Subject: [PATCH 04/14] Update README.md Add support of OpenChat 3.5 --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 3443bb82c15..22b678c5718 100644 --- a/README.md +++ b/README.md @@ -63,6 +63,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **Meditron** | 7B, 70B | | **Mistral** | 7B, 7B-Instruct, 7B-OpenOrca | | **Neural-Chat ** | 7B-v3.1 | +| **OpenChat | 7B-v3.5 | | **OpenLLaMA** | 3B-v2, 7B-v2, 13B-v2 | | **Orca-2** | 7B, 13B | | **Vicuna** | 7B-v1.5, 13B-v1.5 | From 8ee9f5cd5abcfa2a02ddc796e4f287ddfbeeef32 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:33:36 +0100 Subject: [PATCH 05/14] Update models.json bugfix --- api/src/serge/data/models.json | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 6c938618dca..f952b31d5d2 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -8,7 +8,7 @@ "files": [ { "name": "q4_K_M", - "filename": " neural-chat-7b-v3-1.Q4_K_M.gguf", + "filename": "neural-chat-7b-v3-1.Q4_K_M.gguf", "disk_space": 4370000000.0 } ] @@ -24,7 +24,7 @@ "files": [ { "name": "q4_K_M", - "filename": "openlm-research-open_llama_3b_v2-Q4_K_M.gguff", + "filename": "openlm-research-open_llama_3b_v2-Q4_K_M.gguf", "disk_space": 2580000000.0 } ] @@ -62,7 +62,7 @@ "files": [ { "name": "q4_K_M", - "filename": " orca-2-7b.Q4_K_M.gguf", + "filename": "orca-2-7b.Q4_K_M.gguf", "disk_space": 4080000000.0 } ] @@ -73,7 +73,7 @@ "files": [ { "name": "q4_K_M", - "filename": " orca-2-13b.Q4_K_M.gguf", + "filename": "orca-2-13b.Q4_K_M.gguf", "disk_space": 7870000000.0 } ] From 585669d04df37e5c39a76059f3eabd785579e7d3 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:34:31 +0100 Subject: [PATCH 06/14] Update README.md Typo fix --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 22b678c5718..50b84844930 100644 --- a/README.md +++ b/README.md @@ -58,7 +58,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | Category | Models | |:-------------:|:-------| | **CodeLLaMA** | 7B, 13B | -| **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct, 180B, 180b-Chat | +| **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct, 180B, 180B-Chat | | **LLaMA** | 7B, 13B, 70B | | **Meditron** | 7B, 70B | | **Mistral** | 7B, 7B-Instruct, 7B-OpenOrca | From 30dd6712d65b28ab05da213ba24b1770ea21c6cd Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:39:17 +0100 Subject: [PATCH 07/14] Update models.json Add support of Falcon 180B, Falcon 180B-Chat Add support of OpenChat 3.5 --- api/src/serge/data/models.json | 40 +++++++++++++++++++++++++++++++++- 1 file changed, 39 insertions(+), 1 deletion(-) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index f952b31d5d2..17638405363 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -14,7 +14,23 @@ ] } ] - }, + }, + { + "name": "OpenChat", + "models": [ + { + "name": "OpenChat-3_5-7B", + "repo": "TheBloke/openchat_3.5-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "openchat_3.5.Q4_K_M.gguf", + "disk_space": 4370000000.0 + } + ] + } + ] + }, { "name": "Open_LLaMA", "models": [ @@ -110,6 +126,28 @@ { "name": "Falcon", "models": [ + { + "name": "Falcon-180B", + "repo": "TheBloke/Falcon-180B-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "falcon-180b.Q4_K_M.gguf", + "disk_space": 108480000000.0 + } + ] + }, + { + "name": "Falcon-180B-Chat", + "repo": "TheBloke/Falcon-180B-Chat-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": "falcon-180b-chat.Q4_K_M.gguf", + "disk_space": 108480000000.0 + } + ] + }, { "name": "Falcon-40b", "repo": "maddes8cht/tiiuae-falcon-40b-gguf", From 3d065177c005f80a1da1346b3d9dc51ea129b7e0 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:40:31 +0100 Subject: [PATCH 08/14] Update README.md bugfix --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 50b84844930..574c0829d2b 100644 --- a/README.md +++ b/README.md @@ -62,8 +62,8 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **LLaMA** | 7B, 13B, 70B | | **Meditron** | 7B, 70B | | **Mistral** | 7B, 7B-Instruct, 7B-OpenOrca | -| **Neural-Chat ** | 7B-v3.1 | -| **OpenChat | 7B-v3.5 | +| **Neural-Chat** | 7B-v3.1 | +| **OpenChat** | 7B-v3.5 | | **OpenLLaMA** | 3B-v2, 7B-v2, 13B-v2 | | **Orca-2** | 7B, 13B | | **Vicuna** | 7B-v1.5, 13B-v1.5 | From 4f4cc61d8ed1818451e6becd4d2d05543cdc612a Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:44:22 +0100 Subject: [PATCH 09/14] Update models.json neural bugfix --- api/src/serge/data/models.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 17638405363..98ad710f1b6 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -4,7 +4,7 @@ "models": [ { "name": "Neural-Chat-7B-v3-1", - "repo": "TheBloke/neural-chat-7B-v3-1-GGUFF", + "repo": "TheBloke/neural-chat-7B-v3-1-GGUF", "files": [ { "name": "q4_K_M", From 82e5258dc27b6fa92121919533ee6a8e1adc18f6 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:52:39 +0100 Subject: [PATCH 10/14] Update models.json remove Falcon 180B and 180B-Chat --- api/src/serge/data/models.json | 22 ---------------------- 1 file changed, 22 deletions(-) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index 98ad710f1b6..a5ab419764b 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -126,28 +126,6 @@ { "name": "Falcon", "models": [ - { - "name": "Falcon-180B", - "repo": "TheBloke/Falcon-180B-GGUF", - "files": [ - { - "name": "q4_K_M", - "filename": "falcon-180b.Q4_K_M.gguf", - "disk_space": 108480000000.0 - } - ] - }, - { - "name": "Falcon-180B-Chat", - "repo": "TheBloke/Falcon-180B-Chat-GGUF", - "files": [ - { - "name": "q4_K_M", - "filename": "falcon-180b-chat.Q4_K_M.gguf", - "disk_space": 108480000000.0 - } - ] - }, { "name": "Falcon-40b", "repo": "maddes8cht/tiiuae-falcon-40b-gguf", From 3e397546d99cf632a1ebc17c10d23542af89cdef Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Fri, 1 Dec 2023 23:59:15 +0100 Subject: [PATCH 11/14] Update README.md remove Falcon 180B and 180B-Chat --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 574c0829d2b..24859e670dc 100644 --- a/README.md +++ b/README.md @@ -58,7 +58,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | Category | Models | |:-------------:|:-------| | **CodeLLaMA** | 7B, 13B | -| **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct, 180B, 180B-Chat | +| **Falcon** | 7B, 7B-Instruct, 40B, 40B-Instruct | | **LLaMA** | 7B, 13B, 70B | | **Meditron** | 7B, 70B | | **Mistral** | 7B, 7B-Instruct, 7B-OpenOrca | From 62d1e75bf54f74a0cd26da3394839a8482f4063e Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Sat, 2 Dec 2023 01:01:32 +0100 Subject: [PATCH 12/14] Update models.json Add support for Starling-LM-7B-Alpha --- api/src/serge/data/models.json | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index a5ab419764b..f331c137c87 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -1,4 +1,20 @@ [ + { + "name": "Starling", + "models": [ + { + "name": "Starling-LM-7B-Alpha", + "repo": "TheBloke/Starling-LM-7B-alpha-GGUF", + "files": [ + { + "name": "q4_K_M", + "filename": " starling-lm-7b-alpha.Q4_K_M.gguf", + "disk_space": 4370000000.0 + } + ] + } + ] + }, { "name": "Neural-Chat", "models": [ From cb51217ec532ddf231cb2b05b394d66d6983632f Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Sat, 2 Dec 2023 01:03:07 +0100 Subject: [PATCH 13/14] Update README.md Add support for Starling-LM-7B-Alpha --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 24859e670dc..92ba2b1607a 100644 --- a/README.md +++ b/README.md @@ -66,6 +66,7 @@ Instructions for setting up Serge on Kubernetes can be found in the [wiki](https | **OpenChat** | 7B-v3.5 | | **OpenLLaMA** | 3B-v2, 7B-v2, 13B-v2 | | **Orca-2** | 7B, 13B | +| **Starling-LM** | 7B-Alpha | | **Vicuna** | 7B-v1.5, 13B-v1.5 | | **Zephyr** | 7B-Alpha, 7B-Beta | From ca8a34a53b7d7db1981d641c9b4b3484d4ec9432 Mon Sep 17 00:00:00 2001 From: Olivier DEBAUCHE Date: Sat, 2 Dec 2023 01:20:21 +0100 Subject: [PATCH 14/14] Update models.json bugfix --- api/src/serge/data/models.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/api/src/serge/data/models.json b/api/src/serge/data/models.json index f331c137c87..bcb81b44e78 100644 --- a/api/src/serge/data/models.json +++ b/api/src/serge/data/models.json @@ -8,7 +8,7 @@ "files": [ { "name": "q4_K_M", - "filename": " starling-lm-7b-alpha.Q4_K_M.gguf", + "filename": "starling-lm-7b-alpha.Q4_K_M.gguf", "disk_space": 4370000000.0 } ]