# Slow Chat api responses ------ **URL:** \ **Category:** API\ **Created:** [March 13, 2023, 4:49pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171 "2023-03-13T16:49:03Z")\ **Posts on this page:** 18\ **Page:** 1
**Author:** ![Gaouzief](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/gaouzief/32/16161_2.png) [@Gaouzief](https://community.openai.com/u/Gaouzief)\ **Post date:** [March 13, 2023, 4:49pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/1 "2023-03-13T16:49:03Z")
Hello, anyone else experiencing much slower API responses for the chat models for identical calls ? (vs a week ago) Edit: identical prompt on playground is as fast as it used to be, H ---
**Author:** ![samstitt](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/samstitt/32/45006_2.png) [@samstitt](https://community.openai.com/u/samstitt)\ **Post date:** [March 14, 2023, 4:22am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/2 "2023-03-14T04:22:14Z")
Yes, I have apps running identical code to a week ago that are now around 5x as slow. API calls that used to take 3-5s now take 20-30s. I’ve tried playing with max\_tokens and models but this slowdown is consistent. I am using python and very simple API calls. I am assuming this is due to increased usage of the API but would love to know if there is a better solution, or work being done to improve this. Apps which made sense with a fast response (e.g. chat based) are not usable with this slow response time. ---
**Author:** ![nguyenanhdon.qn](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/nguyenanhdon.qn/32/45805_2.png) [@nguyenanhdon.qn](https://community.openai.com/u/nguyenanhdon.qn)\ **Post date:** [March 14, 2023, 4:36am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/3 "2023-03-14T04:36:23Z")
now, 14s for req. If it only takes 3-5 seconds then I can get more benefits, so this is an important issue ---
**Author:** ![ruby\_coder](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ruby_coder/32/19875_2.png) [@ruby\_coder](https://community.openai.com/u/ruby_coder)\ **Post date:** [March 14, 2023, 9:40am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/4 "2023-03-14T09:40:08Z")
I think many often people confuse network delays and data center congestion, etc with API performance. For example, I am 12 time zones away from the US and call the OpenAI completion API, and here are the results when I time the call: #### `text-davinci-003` ```auto Test 1: Completions.get_reply Time: 1.247792 secs Test 2: Completions.get_reply Time: 5.038783 secs Test 3: Completions.get_reply Time: 1.289555 secs Test 4: Completions.get_reply Time: 2.205132 secs ``` Kindly keep in mind that I am testing OpenAI APIs from the opposite side of the world than the US. Also, if I repeat for other models, the results are similar. It’s mostly network traffic issues, not model issues, from my experience. Having said that, lately I have noticed that `text-davinci-002` is about 0.5 seconds faster than `text-davinci-003` (for the same prompt), but did not test extensively. 🙂 ### Appendix: Example Test ![Screenshot 2023-03-14 at 3.51.50 PM](https://us1.discourse-cdn.com/openai1/original/3X/9/2/92b9df3be735c619001967e8923a2b6396e2f28e.png) ---
**Author:** ![Gaouzief](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/gaouzief/32/16161_2.png) [@Gaouzief](https://community.openai.com/u/Gaouzief)\ **Post date:** [March 14, 2023, 10:00am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/5 "2023-03-14T10:00:56Z")
nope, not confusing anything here, again, I see consistant results from api calls made from two servers, Western Europe and Central US, the service is degraded in both locations compared to a week ago, your example uses davinci completions, we are experiencing these issues with chat models ---
**Author:** ![ruby\_coder](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ruby_coder/32/19875_2.png) [@ruby\_coder](https://community.openai.com/u/ruby_coder)\ **Post date:** [March 14, 2023, 10:30am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/6 "2023-03-14T10:30:06Z")
> [@Gaouzief](#): > > your example uses davinci completions, we are experiencing these issues with chat models No problemo… can easily run tests for the `chat completion`: Here are some test results (just now) for `turbo`: ### ` gpt-3.5-turbo-0301` ```auto Test 1, Completion API Time: 1.529 seconds Test 2. Completion API Time: 2.504 seconds Test 3. Completion API Time: 1.557 seconds Test 4. Completion API Time: 1.513 seconds Test 5. Completion API Time: 1.505 seconds ``` ### Appendix: Sample Chat Completion with Time ![Screenshot 2023-03-14 at 5.28.47 PM](https://us1.discourse-cdn.com/openai1/original/3X/6/a/6a49f0925616859394c7e3736bc52666ed9057ea.png) HTH 🙂 ---
**Author:** ![Gaouzief](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/gaouzief/32/16161_2.png) [@Gaouzief](https://community.openai.com/u/Gaouzief)\ **Post date:** [March 14, 2023, 10:32am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/7 "2023-03-14T10:32:28Z")
we are definitely not running a “Hello world” completion… those results are meaning less in the real world, we have a complex prompt, and again, we have seen latency go from 5-8s to above 40… not sure what you are trying to prove 🙂 ---
**Author:** ![Gaouzief](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/gaouzief/32/16161_2.png) [@Gaouzief](https://community.openai.com/u/Gaouzief)\ **Post date:** [March 14, 2023, 10:46am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/9 "2023-03-14T10:46:53Z")
and that’s nowhere near the complexity we have in our prompt, unfortunately I can’t share it. anyway, we were ok with 5-8s, given the task now we are. at 40-50s with some requests above 1mn, which obviously is not production level anymore… hopefully it’s a transient situation ---
**Author:** ![sanjaymk](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/sanjaymk/32/24603_2.png) [@sanjaymk](https://community.openai.com/u/sanjaymk)\ **Post date:** [March 14, 2023, 4:13pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/13 "2023-03-14T16:13:04Z")
I have the same observations - the ChatGPT API has significantly slowed down today. In the env below, ALL my completions were usually in the 3-5 secs range (slow ie). But they are consistently timing out after ~30 secs today. Model: gpt-3.5-turbo servers: US East Coast ---
**Author:** ![rxliuli](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/rxliuli/32/45301_2.png) [@rxliuli](https://community.openai.com/u/rxliuli)\ **Post date:** [March 14, 2023, 4:17pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/14 "2023-03-14T16:17:22Z")
Starting from today, I’ve noticed that the chatgpt API requests are very slow. Even a simple request takes about 13 seconds before returning the first byte. It’s very strange. ![image](https://us1.discourse-cdn.com/openai1/original/3X/6/f/6f0cc004f8bc04ebbcc33f68b21349bf3e033ee0.png) Model: gpt-3.5-turbo Servers: Canada Central ---
**Author:** ![AgusPG](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/aguspg/32/41789_2.png) [@AgusPG](https://community.openai.com/u/AgusPG)\ **Post date:** [March 14, 2023, 4:19pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/15 "2023-03-14T16:19:56Z")
For the folks that are suffering this issue: have you guys also noticed a degradation in the _quality_ of the responses (better if you have specific kpis that prove it)? Or is it just latency? ---
**Author:** ![gaoliqiang1990](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/gaoliqiang1990/32/46834_2.png) [@gaoliqiang1990](https://community.openai.com/u/gaoliqiang1990)\ **Post date:** [March 16, 2023, 4:20am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/16 "2023-03-16T04:20:05Z")
any solutions about the extreme slow speed? ---
**Author:** ![ruby\_coder](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ruby_coder/32/19875_2.png) [@ruby\_coder](https://community.openai.com/u/ruby_coder)\ **Post date:** [March 16, 2023, 4:24am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/17 "2023-03-16T04:24:52Z")
If it meets your use case, you can switch to another model, for example `test-davinci-003` or even better, `text-davinci-002`, for faster response times. HTH 🙂 ---
**Author:** ![moradian.ali](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/moradian.ali/32/22023_2.png) [@moradian.ali](https://community.openai.com/u/moradian.ali)\ **Post date:** [March 20, 2023, 6:58pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/18 "2023-03-20T18:58:00Z")
any update on this ? we’re making chat completion api call openai.createChatCompletion, the same api call took 1s, 4s, 20s when called over and over. ---
**Author:** ![piper](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/piper/32/71261_2.png) [@piper](https://community.openai.com/u/piper)\ **Post date:** [April 12, 2023, 10:54pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/19 "2023-04-12T22:54:21Z")
Has anyone experimented with reducing the size of the prompts? I remember I read it somewhere said the API response time could be related to prompt length. My limited experiment suggests it’s not related, but it’s not conclusive. I don’t know if I send in the same query with long prompt then with short prompt again, are the two response time comparable? Maybe the the second query reused something from the first query? My understanding is that the two queries are totally unrelated on from the API’s perspective. But I’m not sure. ---
**Author:** ![BrianLovesAI](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/brianlovesai/32/145226_2.png) [@BrianLovesAI](https://community.openai.com/u/BrianLovesAI)\ **Post date:** [April 26, 2023, 9:19pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/20 "2023-04-26T21:19:50Z")
![image](https://us1.discourse-cdn.com/openai1/original/3X/7/3/730fedd7b741e6c03851a6a2821fa8609d8ec9c4.png) 😵‍💫 ![image](https://us1.discourse-cdn.com/openai1/original/3X/b/a/ba767f183a4589bfd15841f49c9d4900f3986ad0.png) 😵 I hope… I can solve this speed issue… one day… ---
**Author:** ![yrasowsky](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/yrasowsky/32/34575_2.png) [@yrasowsky](https://community.openai.com/u/yrasowsky)\ **Post date:** [May 2, 2023, 2:56pm UTC](https://community.openai.com/t/slow-chat-api-responses/97171/21 "2023-05-02T14:56:09Z")
So I have been using your suggestion to use da Vinci instead of chat but is there any update to improve the speed of chat? Thank you so much for your valuable time and assistance! ---
**Author:** ![EricGT](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ericgt/32/20571_2.png) [@EricGT](https://community.openai.com/u/EricGT)\ **Post date:** [December 24, 2023, 7:17am UTC](https://community.openai.com/t/slow-chat-api-responses/97171/22 "2023-12-24T07:17:16Z")