OpenAI: Streaming is now available in the Assistants API

https://platform.openai.com/docs/assistants/overview/step-4-create-a-run?context=with-streaming

by jonbraun • 2 years ago

135 86 2 years ago

platform.openai.com

ProjectArcturis• 2 years ago

Has anyone put out a voice-to-text interface for OpenAI? Or anything in the Ollama-verse?

the_newest• 2 years ago

There’s whisper.

eightysixfour2• 2 years ago

OpenAI has a voice to text interface for OpenAI…

willsmith724• 2 years ago

The mobile app is pretty good

Horrendous in non english languages though, the accents are extremely American

notRobot• 2 years ago

Is there a way to use the mobile app on PCs?

I tried with Windows Subsystem for Android but the app refused to work.

arthurcolle5• 2 years ago

Sigh another week lost to the void

nextworddev• 2 years ago

Elaborate?

castles• 2 years ago

"YET ANOTHER shiny new toy to distract me. Can't help myself even though I think it's mostly a waste of time"

Am I just projecting? Relatable, in any case :)

andher• 2 years ago

Finally! I've been using the assistants api in building an ai mock interviewer (https://comp.lol) but the responses were painfully slow when using the latest iterations of the gpt-4 model. This will make things so much more responsive

cosmotic• 2 years ago

I'd still want to see the entire response all at once. Having it stream in while I read it would be very distracting and make it difficult for me to read.

TowerTall• 2 years ago

yes, it like surfing porn in the early internet year using a dialup modem. One line a the time until you finally can see enough of the picture (reply) to realize that is was not the reply you were looking for.

LLM streaming must be a cost saving feature to prevent you from overloading the servers by asking to many questions with in a short time frame. Annoying feature IMHO

qwertox• 2 years ago

It's a request the front-end developer should be confronted with, not OpenAI.

The website could as well buffer the incoming stream until the used clicks an area to request the display of the next block of the response, once he has finished reading the initial sentences.

pieterhg• 2 years ago

Same it was super slow and unusable when I tried. 10 seconds for a reply or smth. GPT4 API itself was way faster

m-p-3• 2 years ago

I thought this was about making the OpenAI app available as a digital assistant on Android, as a replacement to Google.

Oh well..

XCSme3• 2 years ago

Any way to have a consistent system prompt across queries without sending it (and using tokens) for each completion?

eightysixfour2• 2 years ago

The Assistant API handles that, it has the system prompt as part of the assistant that you interact with.

XCSme3• 2 years ago

And can you share the assistant with other users?

Also, the system prompt in assistants doesn't consume tokens?

arthurcolle5• 2 years ago

The assistant has its own "instructions" (replacement for system prompt)

and then on each run, you have the option to add more guidance to the run explicitly, without modifying the assistant instructions (system prompt)

It's a little bit different but kind of the same

baobabKoodaa2• 2 years ago

No, adding run instructions will replace existing instructions for that run

AgentME2• 2 years ago

This was one of the limitations of the Assistants API that made me entirely ignore it up until now.

I am curious if the Assistants API lets you edit/remove/retry messages yet. I don't see anything implying this has changed. It's annoying that the Assistants API doesn't give you enough control to support basic things that the ChatGPT app does.

xvector• 2 years ago

Edit/remove/retry is just including the whole conversation over again (IIUC this is even how the app works.) It's part of why the API is so expensive

AgentME2• 2 years ago

The Assistants API doesn't let you recreate the conversation (with edits or not) because you can't (re)create messages with role=assistant.

varenc• 2 years ago

Like the other commenter said, edit/remove/retry messages can be implemented by the API client already. The API doesn't maintain state so every new message in a "conversation" includes previous messages as context. To edit a message you would re-submit the conversation history with the desired changes.

I get what you're asking for though. It would be nice if this was easier. But that would require OpenAI changing their API model to one where conversation history is stored on their server. It would be more of a "ChatGPT conversation API" then just an GPT-4/3.5 API.

blackoil• 2 years ago

That is what "assistant api" is, you create a thread and add new user message to the thread. The messages are stored on the server.

There is an API to modify messages, though I am not sure of its constraints.

potsandpans• 2 years ago

Openai banned my account for suspicious payment activities, and I never was able to talk to a real person. Just several layers of chat bots posing as people.

I literally want to give them my money and can't. Every few weeks for shirts and giggles i send an email to them saying, "any update on this?"

GaggiX• 2 years ago

I guess it's time for Claude 3 (I imagine you were using it for the LLMs).

thejohnconway• 2 years ago

My Anthropic account was suspended for suspicious activity, even though I never used it. I had forgotten I had signed up, and tried to sign up using a new email with the same phone number. Locked out forever.

slimsag• 2 years ago

Welcome to the future. You might be able to get an enterprise sales contract with human support.

dbish• 2 years ago

One of the reasons I tend to use any of their options through Azure where available. Azure support has a more straight forward (though still sometimes slow) process for account issues.

ukuina• 2 years ago

I suspected as much when one of their support "personnel" used the phrase "I apologize for the earlier confusion..." (there was no confusion, I was simply contradicting what they were saying)

simonw• 2 years ago

Throwing a feature request in here just in case someone from OpenAI sees it.

I'd really like it if the streaming versions of their APIs could return a token usage count at the end.

The non-streaming APIs do this right now:

    curl https://api.openai.com/v1/chat/completions \
      -H "Content-Type: application/json" \
      -H "Authorization: Bearer $OPENAI_API_KEY" -d '{
        "model": "gpt-3.5-turbo",
        "messages": [
          {
            "role": "user",
            "content": "A short fun fact about pigeons"
          }
        ]
      }'

Returns:

    {
      "id": "chatcmpl-92UiIWQaf442wq7Eyp7kF8ge0e3fE",
      "object": "chat.completion",
      "created": 1710381746,
      "model": "gpt-3.5-turbo-0125",
      "choices": [
        {
          "index": 0,
          "message": {
            "role": "assistant",
            "content": "Pigeons are one of the few bird species that can drink water by sucking it up through their beaks, rather than tilting their heads back to swallow."
          },
          "logprobs": null,
          "finish_reason": "stop"
        }
      ],
      "usage": {
        "prompt_tokens": 14,
        "completion_tokens": 33,
        "total_tokens": 47
      },
      "system_fingerprint": "fp_4f0b692a78"
    }

Note the "usage" block there telling me how many tokens were used (which tells me how much this cost).

But if I add "stream": true I get back an SSE stream that looks like this:

    ...
    data: {"id":"chatcmpl-92Uk81oNjrcUJQnPX8fSNqFINLfSI","object":"chat.completion.chunk","created":1710381860,"model":"gpt-3.5-turbo-0125","system_fingerprint":"fp_4f0b692a78","choices":[{"index":0,"delta":{"content":"."},"logprobs":null,"finish_reason":null}]}
    
    data: {"id":"chatcmpl-92Uk81oNjrcUJQnPX8fSNqFINLfSI","object":"chat.completion.chunk","created":1710381860,"model":"gpt-3.5-turbo-0125","system_fingerprint":"fp_4f0b692a78","choices":[{"index":0,"delta":{},"logprobs":null,"finish_reason":"stop"}]}
    
    data: [DONE]

There's no "usage" block, which means I have to try and account for the tokens myself. This is really inconvenient!

I noticed the other day that the Claude streaming API returns a "usage" block with the last message. I'd love it if OpenAI's API did the same thing.

I need this right now because I'm starting to build features for end users of my own software, and I want to be able to give them X,000 tokens "free" before starting to charge them for extras. Counting those tokens myself (probably using tiktoken) is code I'd rather not have to write - especially since features like tools/functions or images make counting tokens a lot less obvious.

harrisonjackson• 2 years ago

This and/or being able to fetch the responses with their token usage by id. What is that ID for without a way to retrieve the completions with it?

gtoubassi• 2 years ago

We do the token counting on our end literally just running tiktoken on the content chunks (although I think usually its one token per chunk). Its a bit annoying and I too expected they'd have the usage block but its one line of code if you already have tiktoken available. I've found the accounting on my side lines up well with what we see on our usage dashboard.

tristanz• 2 years ago

As an FYI, this is fine for rough usage, but it's not accurate. The OpenAI APIs inject various tokens you are unaware of into the input for things like function calling.

megous2• 2 years ago

This website is now like 30% about this probability based autocomplete nonsense. Feels like all those bitcoin hypes and "running everything on blockchain" fad of few years ago. Now it's running everything through "large autocomplete" model.

I really hope this will fade and focus will turn back to highlighting some broader actual human ingenuity in IT, rather than constant stream of "we used autocomplete for this new thing" or "we build this new API for this glorified autocomplete".

Boring.

XCSme3• 2 years ago

I just added this "autocomplete" in my app, and customers emailed to say they actually love it: https://docs.uxwizz.com/guides/ask-ai-new

kfajdsl• 2 years ago

I've gotten so used to having an LLM integrated into my editor that when I work on the occasional spreadsheet (or really anything with syntax that I only use occasionally and no integrated AI) it's pretty jarring to have to go to another tab to look up what function to use for a formula (even if that other tab is ChatGPT).

megous2• 2 years ago

Yes, customers will love anything that helps them. You can get customers to love you by adding any kind of automation for stuff they had to do by hand up to that point. Does this mean there should be 10 articles per day shared about "I added XLSX import to my app, so my customers don't have to do data entry via dialogs"?

My point is about repetitiveness of LLM topics. Not about usefullness of LLM itself. And LLMs are glorified autocomplete. Their internals are maybe interesting, but that's often not what's being discussed here or even written about in the shared articles.

chaxor• 2 years ago

"old man yells at cloud"

Seriously though, it's not going away no matter how much anyone hates it. Emails and blogs will continue to be written with it, letters of recommendation will be/are written with it, Presidential speeches will be written with it, academic articles will be / are written with it (almost all ml and cs research is), news is written with it... It's not going to stop, but it will _probably_/_very likely_ get better.

There is no tool, no human, no method to determine if text is generated with one of these models at high F-score (only sometimes high precision, low recall domains for silly examples).

We're stuck with it. Like the English teacher and their despised spell check.

romanhn• 2 years ago

It occurs to me that over time, reading comprehension will become significantly more important than the ability to write. Anyone will be able to write something smart-sounding with AI's help, but it'll take real skill to make sure the output is correct and appropriate.

xcv1232• 2 years ago

If deep learning algorithms are "autocomplete" then so is the human mind when it strings words together. No, that's not how it works.

dns_snek• 2 years ago

[citation needed]

Just because that makes for a nice narrative in the copyright infringement argument, doesn't make it so.

We know next to nothing about how the human brain works.

ametrau2• 2 years ago

Nah it's got legs as a google replacement / competitor if they keep costs lower and take a smaller rent. WHEN they start advertising they'll explode. Which is why google is trying to snuff them out in the cradle (sorry about the visual).

pedrovhb2• 2 years ago

For all the brilliance in the AI and infra departments of OpenAI, their official Python library (which is the flagship one as I understand) feels pretty unidiomatic, designed without much thought for common patterns in the language.

2012 JavaScript called, it wants its callbacks wrapped in objects back. Why do we have a context manager named "stream" for which you call `.until_done()`? This could've been an iterator, or better - an asynchronous iterator, since this is streaming over the network. We could be destructing instances of named tuples with pattern matching, or even just doing `"".join(delta.text for delta in prompt (...)`. But no here subclass this instead, tells me the wrapper around a web API.

willsmith724• 2 years ago

Everything feels unidiomatic. The API design is bad, the frontends they build are horrific, reliability and availability are shocking.

And yet the AI is so good I put up with them everyday

If they ever grow into a proper product org they'll be unstoppable.

mvkel• 2 years ago

...except for all the others.

Use Claude in Safari and the browser completely locks up after a single response.

athyuttamre• 2 years ago

Hi there, I help design the OpenAI APIs. Would you be able to share more?

You can reply here or email me at atty@openai.com.

(Please don't hold back; we would love to hear the pain points so we can fix them.)

jilles• 2 years ago

Probably written by GPT4

dgellow• 2 years ago

It’s not the case. The SDK is a collaboration between OpenAI and Stainless.

https://www.stainlessapi.com/

As a Stainless contributor I can guarantee you a lot of thoughts has been put into the design, and it definitely isn’t written by an ML model

doctorpangloss• 2 years ago

My experience is their official Python library was easy to use, no surprises, everything is typed and generated from the OpenAPI spec in a thoughtful way.

The tools are great because they don't invent their own DSL, they "just" use JSON schemas.

Maybe they ought to contribute changes to OpenAPI to support streaming APIs better.

In contrast so many startups make their own annotation-driven DSLs for Python with their branding slapped over everything. It gives desperate-for-lock-in vibes. The last people OpenAI should be taking advice from for their API design is this forum.

pedrovhb2• 2 years ago

How is suggesting the use of iterators and named tuples related to creating domain specific languages? If anything I'd say they're a much more generic and universally recognizable approach than having users subclass `AssistantEventHandler` to be passed to `client.beta.threads.runs.create_and_stream`, the context manager. This is very much a long way past just using JSON schemas but that part is ok - there's a REST API, and there's a library. If you're keen on the simplicity of JSON schema then by all means use the API with `requests` or your preferred http client library. Since that's always an option, it stands to reason that the point of having a dedicated library is to provide thoughtful abstractions that make it easier to use the service.

What I'm arguing is precisely that the abstractions in the library (such as the `AssistantEventHandler` shown in the article) are ineffective in making things simpler. They force you to over-engineer solutions and distribute state unnecessarily and be aware of that specific class interface when it could've just been something you use in a `for x in y` loop like everyone would know to do without spending an afternoon looking over docs and figuring out how the underlying implicit FSM works.

rattray3• 2 years ago

Hey there, I helped design the Python library.

The `stream` context manager actually does expose an async iterator (in the async client), so you could instead do this for the simple case:

    with client.beta.threads.runs.create_and_stream(…) as stream:
      async for text in stream.text_deltas:
        print(text, end="", flush=True)

which I think is roughly what you want.

Perhaps the docs should be updated to highlight this simple case earlier.

We are also considering expanding this design, and perhaps replacing the callbacks, like so:

    with client.beta.threads.runs.create_and_stream(…) as stream:
      async for event in stream.all_events:
        if event.type == 'text_delta':
          print(event.delta.value, end='')
        elif event.type == 'run_step_delta':
          event.snapshot.id
          event.delta.step_details...

which I think is also more in line with what you expect. (you could also `match event: case TextDelta: …`).

Note that the context manager is required because otherwise there's no way to tell if you `break` out of the loop (or otherwise stop listening to the stream) which means we can't close the request (and you both keep burning tokens and leak resources in your app).

ametrau2• 2 years ago

Context managers are a great abstraction.

jerrygoyal• 2 years ago

I am interested to use the assistant api for my commercial project but it is not clear from the article what the token count looks like?

- is it counted for a single user message or the sum of all previous messages?

- if there's a file, will it be counted every time a user interacts or only the first time?

visarga• 2 years ago

I think

- it is correlated to the sum, every new interaction adds the whole history again

- yes, but you probably pay for the retrieved fragments, not the whole file

brandall10• 2 years ago

On the second point, there was an issue on launch where it would not find a relevant fragment and appear to load the whole file into the context. Unsure if this has changed but it freaked quite a few folks out OpenAI discussion forums w/ escalating costs.

milar• 2 years ago

Has tool use accuracy improved?

zerop• 2 years ago

they should do streaming for voice inputs on the chatgpt app. right now it's very slow. Voice interfaces need to be streaming

mrtksn3• 2 years ago

Assistant API is too much of a beta still.

I was about to release an app based on the new Assistant API but just a day before the release the response times increased to 8s flat. When I have function calls, that meant up to a minute to get a response.

I had to dismantle everything Assistant API and implement it with Chat API. Which turned out to be great because in Assistant API the context management was very bad and after a few back and forth messages the cost ballooned to over 10K tokens per message.

When I looked closely at the Assistant API and Chat API, I noticed that Assistant API is just a wrapper over Chat API and acts as a web service that stores the previous messages(so slow response problem was probably due to the web server which keeps track of the context). So I went ahead and implemented my own Assistant API which has more control. For example, I set max token cost per message and if the context balloons over that, I make a request with the context and ask OpenAI to create a summary with all the facts so far, add that summary as a system prompt and my context gets compressed back into reasonable territory.

infecto3• 2 years ago

I still don't even know what the Assistant API is supposed to afford me.

mrtksn3• 2 years ago

It's useful if you just need to hook up a chat assistant and don't want to bother with the busywork doing it. All you care is loading the messages from the thread(which are conveniently kept for you) and add new messages.

SoulAuctioneer2• 2 years ago

It does considerably more than (poorly) managing the context window. It also (poorly) enables persistent document storage, knowledge retrieval, function calling and code execution.

johnfurneaux• 2 years ago

Adore. Congrats team. For us the API is epic. We'd just ask for focus on performance.

bytemonitor• 2 years ago

Thanks for posting. I got an example working with functions and tool_calls if anyone needs it. I could not find good examples in the docs. https://medium.com/@hawkflow.ai/openai-streaming-assistants-...

OpenAI: Streaming is now available in the Assistants API

Related Stories