nat / openplayground Goto Github PK

View Code? Open in Web Editor NEW

6.1K 6.1K 466.0 962 KB

An LLM playground you can run on your laptop

License: MIT License

TypeScript 64.72% CSS 1.65% HTML 0.30% JavaScript 0.36% Python 32.97%

openplayground's Introduction

👋

openplayground's People

Contributors

Stargazers

Watchers

Forkers

demgiran harish-garg theodorexli mammarai karbon0x llegomark maddyonline zzmjohn meetpateltech nkarpov rubosstech hqman devoredevelops jcmiller kenychen codedcclxxvii lipiji hz-xlnc jawond jonpol01 aadabi bmorphism chrishabiak frostbite1536 jrhizor ashishb mmyjona mtarapata cling975 iamkomen sanlalex nikunjsrashtasoft n1n9-jp kustomzone srashtasoft daniel-nieto osamaishtiaq jiangxue0820 shafiahmed xlee vinhphu3000 hitesh97 kappuchino kaupo-parmas commerceless fujohnwang kylegoddard starmorph kalufinnle ironsentinel gibson20200202 awdedarkar kosukesaigusa matrak ekaone mozekin deniska83 theogbrand standardgalactic riveri1 csaroff caglarkoksal classicrob dev-cj kameleon-ad segmond robertogentile jogfeelingvi shunta-anezaki keeganmccallum mcicolella warehouse-forks mohammadzainabbas gagliardetto cprevot93 devmuhammadasad djstompzone gitntel iacore devbox10 mrwoodsimple telecloud-ca richardsonjf mmizutani francyjglisboa jrcribb mayerchatgpt leeaandrob bitifirefly alejandrosuarez ylcoldplayer helcc sksundaram-learning all-forks woonhock abaso007 fcarokia ravyt3ja jhsygg sbepstein

openplayground's Issues

Ability to add Azure as an openai provider

We use Azure in-house, would love to be able to use them as a provider on the openplayground.

`openplayground -e` is not accepting path to env file

openplayground -e .env will not respect the existing .env file and instead overwrite it with a blank file.

What software was used to make `all-features.mp4`?

As there are no Discussions, this is an "Issue". How was that demo video made? It is too... smooth and perfect to be a screen capture. A good UI showcase indeed.

Add gpt-4-32k model

It would be great to make available the gpt-4-32k model.

Prompt loss when switching tabs

If you write a prompt in the Playground and then switch to compare without hitting Submit, the playground won't save your prompt and it will be lost when you return to the Playground tab.

An attempt was made to access a socket in a way forbidden by its access permissions

When I do openplayground run, this is what I'm getting:

Initializing download manager...
Download loop started...
 * Serving Flask app 'server.app'
 * Debug mode: off
An attempt was made to access a socket in a way forbidden by its access permissions

I ran the cmd prompt with admin and still got the same message.

no models by default in the combo

after laucing pip and run I got the following empty interface

so, is impossible to do anything with this

Ability to add system message in GPT-3.5-turbo?

Just want to confirm this is/is not possible atm. Can't see an option?

First, thank you a lot Nat and everybody who contributed to providing this as a hosted service as well as open-sourcing it.
This is genuinely really cool, especially considering that some of the models are themselves open source.

Would you consider adding a GDPR-compliant privacy policy (even just the IP address is considered personal data)?

That would be much appreciated.

i run the software in local host , i get text-davinchi-3 maximun token is 1024 ! how to make it 4000 ??????

is there an API?

Wondering how I can use this instead of openai api -- does this program provide a rest api?

freezing on linux

The entire OS freezes up for me. Happens on both manjaro and ubuntu (server)

After downloading a few models the os freezes up.

After run calling localhost:5432 there is a white page

After pip and run, in localhost:5432 I have a white page, only title in tab report OpenPlayground.
I open developer tool and in console I have this error message:

Failed to load module
script: Expected a JavaScript module script but the server responded with a MIME type of "text/plain". Strict MIME type checking is enforced for module scripts per HTML spec.

i added hugging face read api key, but when i try to run a query i get a 503

can we get an aur package for arch

openplayground

huggingface models pending

why when i put in my huggingface api key and select a gpt4 model, does it just say pending? what am i supposed to do?

how to run in production mode?

What does it mean "use a wsgi server"??

model links in readme are broken

I'm trying to get the openai chat-gtp working, but I don't know how.

I paid $5 for my another email count

I have paid $5 for my another email count in nat.dev ,but the gmail account. is there any way to refund the money?

add upload for models

Can you please provide an example of how to load a .bin model please?

whats Stop Sequences do ? and how to use it ?

and thr json files i download in pc whats i need from it and how to use ?

Unknown error

i have dolly installed locally and try running a query and it just throws a generic error.

INFO:server.app:Received inference request huggingface-local
INFO:server.lib.inference:Requesting inference from databricks/dolly-v2-12b on huggingface-local
INFO:server.lib.inference:Starting inference for 1 - databricks/dolly-v2-12b

Error

Unknown error

Error building frontend app

Have been running into this error below building the frontend app npx parcel build src/index.html --no-cache. I'm running node version 18.15. Is there any dependency that I misconfigured?

The error message is:

Error: Invalid regular expression: /^[$_\p{ID_Start}][$_\u200C\u200D\p{ID_Continue}]*$/: Invalid property name in character class at /openplayground/app/node_modules/@parcel/packager-js/lib/ScopeHoistingPackager.js:91

I Lost Balance on nat.dev Account after I Delete Account accidentally

@nat
Regrettably, I accidentally deleted my account while attempting to log out. I noticed that there are two buttons on the page, one for deleting the account and the other for logging out, which can be easily confused. When I logged back in, I realized that my balance was missing.

I was wondering if you could help me recover my balance. My Google account associated with this nat.dev account is [email protected]. I would greatly appreciate any assistance you can provide to help resolve this issue.

Help .. increase gpt4 model maximum tokens more than 8192 PLEASE

Help .. increase gpt4 model maximum tokens more than 8192 PLEASE in https://nat.dev/ website . need more than 8192 please ... if can make it 16000

model error gpt4

please help , when i ask for question in chatgpt 4 , in half of answer my question i got pop up message that model error no response from openai after 60 seconds ! please help

Unknown error

为什么我冲了钱，用不了了？

Document API

Add GHA CI/CD Workflow

Curious to know what build jobs you'd like to happen. Assuming these:

Twine/PyPI packaging
flake8 linting
Pytest with coverage report generation (needs tests first :) )
Upload artifact(s)
Download artifact(s) and Deployment?
Security scans?
Release notes/Package releases?
Containerization?

Let me know how I can help.

oasst-llama

can we get llama and vicuna here as well? There's also oasst llama and RWKW.

GPT-4 is GPT-3?

I found an interesting thing, when I asked GPT is GPT- 4model?He answered me that he is GPT - 3. Un....

A UI issue with non-latin characters

The prompt is to translate a given paragraph to Arabic. It produces identical results but the characters are enclosed in a <span> HTML tag which make them extremely hard to read. See picture below.

Your card has been declined. when buy ! Help

please help i try many cards and get same issue Your card has been declined. this error message pop up after click buy .. please help

OpenAI API request was invalid: The model: `gpt-4` does not exist

Add openAI API key

select model gpt-4

enter prompt

press submit

result: OpenAI API request was invalid: The model: gpt-4 does not exist

`openplayground run` results in error

Macbook with M1 Chip

This is the error:

OMP: Error #15: Initializing libiomp5.dylib, but found libiomp5.dylib already initialized.
OMP: Hint This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results. For more information, please see http://www.intel.com/software/products/support/.
fish: Job 1, 'openplayground run' terminated by signal SIGABRT (Abort)

Support for Dolly and Open Assistant

Recently Dolly 2 came out and it's completely open source (not based on Llama). Some models from Open Assistant are already available and the official release is about to happen.

It would be nice to have Dolly 2 and Open Assistant support.

duplicate filename

INFO:server.app:Received inference request huggingface-local
INFO:server.lib.inference:Requesting inference from chavinlo/gpt4-x-alpaca on huggingface-local
INFO:server.lib.inference:Starting inference for 1 - chavinlo/gpt4-x-alpaca
INFO:werkzeug:127.0.0.1 - - [25/Apr/2023 00:46:59] "POST /api/inference/text/stream HTTP/1.1" 200 -
ERROR:server.lib.inference:Error: Couldn't build proto file into descriptor pool: duplicate file name sentencepiece_model.proto
INFO:server.lib.inference:Completed inference for chavinlo/gpt4-x-alpaca on huggingface-local
INFO:server.lib.api.inference:Done streaming SSE

Chat UI too limiting?

Given the existence of ChatGPT and the chat-mode completions OpenAI provides, it makes sense to be able to build a chat interface in openplayground.

However, I wonder if framing this feature as a "chat UI" is too limiting? The chat UI itself is really just a specific implementation of a stateful workflow: "process user input (in the chat case, wrap it in JSON), run completion with some history, process model output (in the chat case, parse JSON and extract/parse the response)."

Without going so far as saying "openplayground needs to be natto.dev", I wonder if there's a middle ground here.

🤔

Disable logprops checkbox if model doesn't support it

my balance has disappeared

I used to have 4 dollars, it just disappeared without usage

gpt-4 option is not the real GPT-4 model?

When I choose the 'gpt-4' model and just ask it 'which model are you? gpt-3 or gpt 4?' It just says that ' the model is gpt 3.5 turbo, the gpt 4 model is not released ' some kind of things, and when I compare the gpt-4 with gpt3.5-turbo, I can't find any conspicuous evidence that the gpt-4 option actually use GPT-4 model. Can somebody explain why?

how run as desktop application?

Hello,
I have tried the web gui and found it quite nice, but I was wondering if it would be possible to run even just one model as an application outside of a web server completely "locally" idk how to explain this better, basically just having the ability to run models with a gui as an application not web server on a computer to get a ChatGPT like experience but completely just outside of the browser like searching in explorer or finder even if thats a terrible example.
Thanks!

Copy-paste of some text doesn't include newlines

Hard line breaks aren't working, e.g. copy-paste this text into the playground:
https://gist.github.com/nat/8b8788b5e4bf8fa3863f80da0c1332c9

And you get this:

Introduce Markdown code block

Put code output in a mark down code block, accompanied by a copy button.

Is this a scam website? I topped up five dollars, used only a few cents, and then my balance was cleared to zero. They are asking me to top up again.

model search is broken

$ openplayground run
INFO:server.lib.sseserver:SUBSCRIBING TO: inferences
INFO:server.lib.sseserver:SUBSCRIBING TO: notifications
INFO:server.lib.sseserver:GETTING TOPIC: notifications
INFO:server.lib.sseserver:GETTING TOPIC: inferences
INFO:server.app:Initializing download manager...
INFO:server.app:Download loop started...
 * Serving Flask app 'server.app'
 * Debug mode: off
INFO:werkzeug:WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
 * Running on http://localhost:5432
INFO:werkzeug:Press CTRL+C to quit
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:20] "GET / HTTP/1.1" 200 -
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:20] "GET /index.d826928d.js HTTP/1.1" 200 -
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:20] "GET /index.e96bf64d.css HTTP/1.1" 200 -
INFO:server.lib.api:Getting enabled models
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:20] "GET /api/models-enabled HTTP/1.1" 200 -
INFO:server.lib.api:Received notification request
INFO:server.lib.sseserver:LISTENING TO: notifications
INFO:server.lib.sseserver:LISTENING
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:20] "GET /favicon.ico HTTP/1.1" 200 -
INFO:server.lib.api:Getting enabled models
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:23] "GET /api/models-enabled HTTP/1.1" 200 -
INFO:server.lib.api:Getting enabled models
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:29] "GET /api/models-enabled HTTP/1.1" 200 -
INFO:server.lib.api:Getting providers with models
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:36] "GET /api/providers-with-key-and-models HTTP/1.1" 200 -
INFO:server.lib.api.provider:Searching Provider Models openplayground
ERROR:server.app:Exception on /api/provider/openplayground/models/search [GET]
Traceback (most recent call last):
  File "/usr/lib/python3.10/site-packages/urllib3/connection.py", line 174, in _new_conn
    conn = connection.create_connection(
  File "/usr/lib/python3.10/site-packages/urllib3/util/connection.py", line 72, in create_connection
    for res in socket.getaddrinfo(host, port, family, socket.SOCK_STREAM):
  File "/usr/lib/python3.10/socket.py", line 955, in getaddrinfo
    for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
socket.gaierror: [Errno -3] Temporary failure in name resolution

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 703, in urlopen
    httplib_response = self._make_request(
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 386, in _make_request
    self._validate_conn(conn)
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 1042, in _validate_conn
    conn.connect()
  File "/usr/lib/python3.10/site-packages/urllib3/connection.py", line 358, in connect
    self.sock = conn = self._new_conn()
  File "/usr/lib/python3.10/site-packages/urllib3/connection.py", line 186, in _new_conn
    raise NewConnectionError(
urllib3.exceptions.NewConnectionError: <urllib3.connection.HTTPSConnection object at 0x7fee4b826bc0>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.10/site-packages/requests/adapters.py", line 489, in send
    resp = conn.urlopen(
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 787, in urlopen
    retries = retries.increment(
  File "/usr/lib/python3.10/site-packages/urllib3/util/retry.py", line 592, in increment
    raise MaxRetryError(_pool, url, error or ResponseError(cause))
urllib3.exceptions.MaxRetryError: HTTPSConnectionPool(host='openplayground.filler', port=443): Max retries exceeded with url: /api/search?q=gpt4 (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fee4b826bc0>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution'))

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 2528, in wsgi_app
    response = self.full_dispatch_request()
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 1825, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask_cors/extension.py", line 165, in wrapped_function
    return cors_after_request(app.make_response(f(*args, **kwargs)))
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 1823, in full_dispatch_request
    rv = self.dispatch_request()
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 1799, in dispatch_request
    return self.ensure_sync(self.view_functions[rule.endpoint])(**view_args)
  File "/home/ettinger/.local/lib/python3.10/site-packages/server/lib/api/provider.py", line 111, in provider_models_search
    response = requests.get(search_url)
  File "/usr/lib/python3.10/site-packages/requests/api.py", line 73, in get
    return request("get", url, params=params, **kwargs)
  File "/usr/lib/python3.10/site-packages/requests/api.py", line 59, in request
    return session.request(method=method, url=url, **kwargs)
  File "/usr/lib/python3.10/site-packages/requests/sessions.py", line 587, in request
    resp = self.send(prep, **send_kwargs)
  File "/usr/lib/python3.10/site-packages/requests/sessions.py", line 701, in send
    r = adapter.send(request, **kwargs)
  File "/usr/lib/python3.10/site-packages/requests/adapters.py", line 565, in send
    raise ConnectionError(e, request=request)
requests.exceptions.ConnectionError: HTTPSConnectionPool(host='openplayground.filler', port=443): Max retries exceeded with url: /api/search?q=gpt4 (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fee4b826bc0>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution'))
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:02:57] "GET /api/provider/openplayground/models/search?query=gpt4 HTTP/1.1" 500 -
INFO:server.lib.api.provider:Searching Provider Models openplayground
ERROR:server.app:Exception on /api/provider/openplayground/models/search [GET]
Traceback (most recent call last):
  File "/usr/lib/python3.10/site-packages/urllib3/connection.py", line 174, in _new_conn
    conn = connection.create_connection(
  File "/usr/lib/python3.10/site-packages/urllib3/util/connection.py", line 72, in create_connection
    for res in socket.getaddrinfo(host, port, family, socket.SOCK_STREAM):
  File "/usr/lib/python3.10/socket.py", line 955, in getaddrinfo
    for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
socket.gaierror: [Errno -3] Temporary failure in name resolution

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 703, in urlopen
    httplib_response = self._make_request(
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 386, in _make_request
    self._validate_conn(conn)
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 1042, in _validate_conn
    conn.connect()
  File "/usr/lib/python3.10/site-packages/urllib3/connection.py", line 358, in connect
    self.sock = conn = self._new_conn()
  File "/usr/lib/python3.10/site-packages/urllib3/connection.py", line 186, in _new_conn
    raise NewConnectionError(
urllib3.exceptions.NewConnectionError: <urllib3.connection.HTTPSConnection object at 0x7fee4b824610>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.10/site-packages/requests/adapters.py", line 489, in send
    resp = conn.urlopen(
  File "/usr/lib/python3.10/site-packages/urllib3/connectionpool.py", line 787, in urlopen
    retries = retries.increment(
  File "/usr/lib/python3.10/site-packages/urllib3/util/retry.py", line 592, in increment
    raise MaxRetryError(_pool, url, error or ResponseError(cause))
urllib3.exceptions.MaxRetryError: HTTPSConnectionPool(host='openplayground.filler', port=443): Max retries exceeded with url: /api/search?q=gpt4 (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fee4b824610>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution'))

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 2528, in wsgi_app
    response = self.full_dispatch_request()
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 1825, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask_cors/extension.py", line 165, in wrapped_function
    return cors_after_request(app.make_response(f(*args, **kwargs)))
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 1823, in full_dispatch_request
    rv = self.dispatch_request()
  File "/home/ettinger/.local/lib/python3.10/site-packages/flask/app.py", line 1799, in dispatch_request
    return self.ensure_sync(self.view_functions[rule.endpoint])(**view_args)
  File "/home/ettinger/.local/lib/python3.10/site-packages/server/lib/api/provider.py", line 111, in provider_models_search
    response = requests.get(search_url)
  File "/usr/lib/python3.10/site-packages/requests/api.py", line 73, in get
    return request("get", url, params=params, **kwargs)
  File "/usr/lib/python3.10/site-packages/requests/api.py", line 59, in request
    return session.request(method=method, url=url, **kwargs)
  File "/usr/lib/python3.10/site-packages/requests/sessions.py", line 587, in request
    resp = self.send(prep, **send_kwargs)
  File "/usr/lib/python3.10/site-packages/requests/sessions.py", line 701, in send
    r = adapter.send(request, **kwargs)
  File "/usr/lib/python3.10/site-packages/requests/adapters.py", line 565, in send
    raise ConnectionError(e, request=request)
requests.exceptions.ConnectionError: HTTPSConnectionPool(host='openplayground.filler', port=443): Max retries exceeded with url: /api/search?q=gpt4 (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fee4b824610>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution'))
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:03:00] "GET /api/provider/openplayground/models/search?query=gpt4 HTTP/1.1" 500 -
INFO:server.lib.api:Getting enabled models
INFO:werkzeug:127.0.0.1 - - [23/Apr/2023 00:03:11] "GET /api/models-enabled HTTP/1.1" 200 -
INFO:server.lib.api:Getting enabled models

该好好

Help - where is server/models.json?

Where is this file?

If I run the server and go to localhost:5432/models.json there's no file to edit there.

Also, are there any templates for the llama models? I mean, I'm grateful for this entire project for free, but it seems kind of strange to have to write the template for each model manually, but I don't know...🤷‍♂️ I don't even know what all the parameters are or their ranges.

my credit has disappeared

I had 5 bucks, it just disappeared without being used

No response from openai after sixty seconds

i got this error every tine when ask for long answer in gpt4 , the error come up after 3 minuites and cut the answer ! please fix it . its cost money