Jared Van Bortel
d8c842263f
python: more fixes for new prompt templates ( #2044 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-06 14:22:08 -05:00
Jared Van Bortel
5a874be7c1
modellist: rename "deprecated" to "removedIn", disable if equal ( #2063 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-06 14:12:21 -05:00
Jared Van Bortel
402f515a5d
chat: fix ChatGPT after #1970 ( #2051 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-06 14:02:18 -05:00
Jared Van Bortel
2a91ffd73f
chatllm: fix undefined behavior in resetContext
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-06 12:54:19 -06:00
Jared Van Bortel
0fc071d228
chat: better handle case where network reachability is unknown
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-06 12:52:37 -06:00
Jared Van Bortel
c19b763e03
llmodel_c: expose fakeReply to the bindings ( #2061 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-06 13:32:24 -05:00
Adam Treat
be6d3bf9dc
Bump version and release notes for 2.7.2
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-03-05 13:15:21 -05:00
Adam Treat
83c76be68a
Model discovery.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-03-05 11:31:47 -05:00
ThiloteE
f2b4809b72
models3: remove system prompt of Nous-Hermes-2-Mistral-7b-DPO ( #2054 )
...
Signed-off-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
2024-03-01 14:19:18 -05:00
Jared Van Bortel
9fafca5c94
qml: update models.json URL in error message
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-03-01 13:50:10 -05:00
Adam Treat
7d1e30766f
Fix the hash on the new model.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-27 09:56:11 -05:00
Adam Treat
5ddcf61ae4
Shorten the description and provide a valid url.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-27 09:34:50 -05:00
ThiloteE
713afb7070
Add-Nous-Hermes-2-Mistral-7B-DPO.Q4_0.gguf
...
Adds Nous-Hermes-2-Mistral-7B-DPO.Q4_0.gguf, which is the new 7b flagship model of NousResearch.
**Original Model location:**
https://huggingface.co/NousResearch/Nous-Hermes-2-Mistral-7B-DPO-GGUF
**Model description:**
Nous Hermes 2 on Mistral 7B DPO is the new flagship 7B Hermes! This model was DPO'd from Teknium/OpenHermes-2.5-Mistral-7B and has improved across the board on all benchmarks tested - AGIEval, BigBench Reasoning, GPT4All, and TruthfulQA.
The model prior to DPO was trained on 1,000,000 instructions/chats of GPT-4 quality or better, primarily synthetic data as well as other high quality datasets, available from the repository teknium/OpenHermes-2.5.
**Original Dataset Location:**
https://huggingface.co/datasets/teknium/OpenHermes-2.5
**Dataset description:**
This is the dataset that made OpenHermes 2.5 and Nous Hermes 2 series of models.
The Open Hermes 2/2.5 and Nous Hermes 2 models have made significant advancements of SOTA LLM's over recent months, and are underpinned by this exact compilation and curation of many open source datasets and custom created synthetic datasets.
The Open Hermes 2.5 dataset is a continuation of the Open Hermes 1 dataset, at a much larger scale, much more diverse, and much higher quality compilation, reaching 1M, primarily synthetically generated instruction and chat samples.
Signed-off-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
2024-02-27 08:28:43 -06:00
Jared Van Bortel
4a16a920a3
python: actually fix python 3.8 compatibility ( #1973 )
...
importlib.resources.files also didn't exist until python 3.9.
Fixes #1972
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-26 13:15:02 -05:00
Jared Van Bortel
a59645c839
python: fix mistakes from PR #1970 ( #2023 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-26 13:11:51 -05:00
Jared Van Bortel
f500bcf6e5
llmodel: default to a blank line between reply and next prompt ( #1996 )
...
Also make some related adjustments to the provided Alpaca-style prompt templates
and system prompts.
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-26 13:11:15 -05:00
Jared Van Bortel
fc1a281381
modellist: fix bad copy-paste in ModelList::clone ( #2011 )
...
s/contextLength/gpuLayers/
Fixes #2010
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-26 13:09:29 -05:00
Jared Van Bortel
007d469034
bert: fix layer norm epsilon value ( #1946 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-26 13:09:01 -05:00
AT
7a23b23728
Update gpt4all-chat/modellist.cpp
...
Co-authored-by: Jared Van Bortel <jared@nomic.ai>
Signed-off-by: AT <manyoso@users.noreply.github.com>
2024-02-26 12:04:16 -06:00
Adam Treat
f720261d46
Fix another vulnerable spot for crashes.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-26 12:04:16 -06:00
Adam Treat
17a2cdbe35
Fix crasher with layer count
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-26 12:04:16 -06:00
Jared Van Bortel
72474a2efa
ci: fix chat installer build by updating QtIFW dependency ( #2015 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-26 11:47:11 -05:00
chrisbarrera
f8b1069a1c
add min_p sampling parameter ( #2014 )
...
Signed-off-by: Christopher Barrera <cb@arda.tx.rr.com>
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
2024-02-24 17:51:34 -05:00
TareHimself
a153cc5b25
typescript: async generator and token stream ( #1897 )
...
Signed-off-by: Tare Ebelo <75279482+TareHimself@users.noreply.github.com>
Signed-off-by: jacob <jacoobes@sern.dev>
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Co-authored-by: jacob <jacoobes@sern.dev>
Co-authored-by: Jared Van Bortel <jared@nomic.ai>
2024-02-24 17:50:14 -05:00
Adam Treat
ef518fae3e
Fix crash with chatgpt and gpu layers.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-22 15:51:56 -06:00
Jared Van Bortel
e7f2ff189f
fix some compilation warnings on macOS
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-22 15:09:06 -05:00
Jared Van Bortel
88e330ef0e
llama.cpp: enable Kompute support for 10 more model arches ( #2005 )
...
These are Baichuan, Bert and Nomic Bert, CodeShell, GPT-2, InternLM,
MiniCPM, Orion, Qwen, and StarCoder.
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-22 14:34:42 -05:00
Jared Van Bortel
fc6c5ea0c7
llama.cpp: gemma: allow offloading the output tensor ( #1997 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-22 14:06:18 -05:00
Jared Van Bortel
c1dcb3f5b8
models.json: fix Mistral OpenOrca filename
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-22 08:57:51 -06:00
Adam Treat
a010a8a7ca
Bump version and release notes for v2.7.1
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 16:54:08 -05:00
Jared Van Bortel
ef0a67eb94
models: remove gemma from models2.json and models3.json ( #1995 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 16:18:26 -05:00
Adam Treat
67bbce43ab
Fix state issues with reloading model.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 16:05:49 -05:00
Jared Van Bortel
4fc4d94be4
fix chat-style prompt templates ( #1970 )
...
Also use a new version of Mistral OpenOrca.
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 15:45:32 -05:00
Jared Van Bortel
b8f5c74f40
add models3.json for new templates ( #1993 )
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 15:41:20 -05:00
Jared Van Bortel
c13202a6f5
models2.json: gemma requires a future GPT4All version
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 14:43:55 -05:00
Jared Van Bortel
4a8c6d7f9c
gemma: fix default prompt template
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 13:36:31 -06:00
Jared Van Bortel
32837fb3a0
models2.json: add gemma model
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 13:36:31 -06:00
Jared Van Bortel
7810b757c9
llamamodel: add gemma model support
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-21 13:36:31 -06:00
Adam Treat
896fc6fbb7
Save the window size for the user and reuse next load.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 11:54:26 -06:00
Adam Treat
fa0a2129dc
Don't try and detect model load error on startup.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Adam Treat
b0c471aed8
Make the reload/regenerate buttons a little bit larger font.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Adam Treat
67099f80ba
Add comment to make this clear.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Adam Treat
ad34c2bdd4
Don't erase context when reloading model by selection.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Adam Treat
fbf5e5e732
Increase padding for elided text in combo.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Adam Treat
ed0f93977d
Fixes for issues identified in review.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Adam Treat
d948a4f2ee
Complete revamp of model loading to allow for more discreet control by
...
the user of the models loading behavior.
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-21 10:15:20 -06:00
Simon Willison
f2024a1f9e
python: README and project links for PyPI listing ( #1964 )
...
Signed-off-by: Simon Willison <swillison@gmail.com>
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Co-authored-by: Jared Van Bortel <jared@nomic.ai>
2024-02-13 17:44:33 -05:00
Jared Van Bortel
6fdec808b2
backend: update llama.cpp for faster state serialization
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-13 17:39:18 -05:00
Jared Van Bortel
a1471becf3
backend: update llama.cpp for Intel GPU blacklist
...
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2024-02-12 13:16:24 -05:00
Adam Treat
16927d9a76
Fix visual artifact with close button in new version dialog.
...
Signed-off-by: Adam Treat <treat.adam@gmail.com>
2024-02-12 12:25:33 -05:00