# GPT-3.5 Turbo fine-tuning now available (and new GPT3 models) **URL:** \ **Category:** API\ **Tags:** announcement, fine-tuning, api\ **Created:** [August 22, 2023, 7:32pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425 "2023-08-22T19:32:01Z")\ **Posts on this page:** 19\ **Page:** 1
**Author:** ![novaphil](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/novaphil/32/130934_2.png) [@novaphil](https://community.openai.com/u/novaphil)\ **Post date:** [August 22, 2023, 7:32pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/1 "2023-08-22T19:32:01Z")
New blog post with announcement: > **[GPT-3.5 Turbo fine-tuning and API updates](https://openai.com/blog/gpt-3-5-turbo-fine-tuning-and-api-updates)** > > Developers can now bring their own data to customize GPT-3.5 Turbo for their use cases. > Fine-tuning for GPT-3.5 Turbo is now available, with fine-tuning for GPT-4 coming this fall. … Early tests have shown a fine-tuned version of GPT-3.5 Turbo can match, or even outperform, base GPT-4-level capabilities on certain narrow tasks. > Fine-tuning with GPT-3.5-Turbo can also handle 4k tokens—double our previous fine-tuned models. Early testers have reduced prompt size by up to 90% by fine-tuning instructions into the model itself, speeding up each API call and cutting costs. > Today, we are making `babbage-002` and `davinci-002` available, either as base or fine-tuned models. Customers can access those models by querying the [Completions API](https://platform.openai.com/docs/api-reference/completions). > These models can be fine-tuned with our new API endpoint `/v1/fine_tuning/jobs` . This new endpoint offers pagination and more extensibility to support the future evolution of the fine-tuning API. Transitioning from `/v1/fine-tunes` to the updated endpoint is straightforward and more details can be found in our new [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning). ---
**Author:** ![anon22939549](https://avatars.discourse-cdn.com/v4/letter/a/f14d63/32.png) [@anon22939549](https://community.openai.com/u/anon22939549)\ **Post date:** [August 22, 2023, 8:22pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/2 "2023-08-22T20:22:47Z")
This is going to be _huge_! Lots of potential for some very cool applications. This is incredibly exciting! ---
**Author:** ![chagonkas](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/chagonkas/32/154217_2.png) [@chagonkas](https://community.openai.com/u/chagonkas)\ **Post date:** [August 22, 2023, 8:42pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/3 "2023-08-22T20:42:11Z")
Does anyone know if you can fine tune for function calls? There is no documentation around it, but was hoping you might be able to do something like this: ```auto "messages": [ { "role": "system", "content": "You are a general purpose agent" }, { "role": "user", "content": "What's the weather in Jackson, wy" }, { "function_call": "get_weather", args: {"location": "jackson, wy"} } ] ``` Then again, I’m not even 100% sure this version of GPT3.5 supports function calls at all… ---
**Author:** ![nunodonato](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/nunodonato/32/7022_2.png) [@nunodonato](https://community.openai.com/u/nunodonato)\ **Post date:** [August 22, 2023, 8:42pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/4 "2023-08-22T20:42:24Z")
I’m really confused about davinci-002. Do they mean the older models? We used to be able to fine-tune davinci-003 ---
**Author:** ![novaphil](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/novaphil/32/130934_2.png) [@novaphil](https://community.openai.com/u/novaphil)\ **Post date:** [August 22, 2023, 8:49pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/5 "2023-08-22T20:49:11Z")
> [@chagonkas](#): > > Does anyone know if you can fine tune for function calls? “Support for fine-tuning with function calling and gpt-3.5-turbo-16k will be coming later this fall.” * * * > [@nunodonato](#): > > I’m really confused about davinci-002. Do they mean the older models? We used to be able to fine-tune davinci-003 You could finetune `davinci`, but not `text-davinci-003` “Fine-tuning is currently only available for the following base models: `davinci` , `curie` , `babbage` , and `ada` . These are the original models that do not have any instruction following training (like `text-davinci-003` does for example).” ---
**Author:** ![\_j](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/_j/32/766292_2.png) [@\_j](https://community.openai.com/u/_j)\ **Post date:** [August 22, 2023, 9:02pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/6 "2023-08-22T21:02:35Z")
Notable: **Can I continue fine-tuning a model that has already been fine-tuned?** ([OpenAI Platform](https://platform.openai.com/docs/guides/fine-tuning/can-i-continue-fine-tuning-a-model-that-has-already-been-fine-tuned)) No, we do not currently support continuing the fine-tuning process once a job has finished. We plan to support this in the near future. **Pricing:** fine-tune `davinci-002` output is **1/10th** the cost of GPT-3 davinci fine-tune - and the same price as curie (without direct replacement). * * * It will be interesting to explore deeper what comes along with the pretrained/reweighted replacement completion models, now available in the playground. Definitely a difference in the output and wildly different runs at temperature and top-p as low as 0.5 with the curiously-behaving babbage-002, with only one result that is close to accurate: **old babbage:** A few-shot training is a technique where the model is trained on a small set of data. The idea is that the model will learn the “rules” of the data, and it will be able to answer questions about the data with a high degree of accuracy. **babbage-002** : (freq-penalty 1) It’s a method of training AI systems that is not based on the use of large amounts of data. Instead, it relies on the use of small amounts of data to train a model. The term “few-shot” refers to the fact that only a small number (typically 10-20) questions are used for training. **babbage-002** : (penalties 0.05) It’s a process where a large number of questions are asked to the AI, and the answers are generated based on the questions that were asked. This process is useful for learning new concepts, and is a good way to get started with machine learning. **babbage-002** : (again) It’s a technique for training an AI to perform well on a small number of test data, and then use that data to improve the performance of the AI on a larger set of test data. Quite high perplexity gives randomness (and also got a loop output): ![image](https://us1.discourse-cdn.com/openai1/original/3X/1/5/155025ecfd6afca5ee9e383626d2000151eb0077.png) * * * and are the new models already fine-tuned for you? **davinci** (base, old) obeys: ![image](https://us1.discourse-cdn.com/openai1/original/3X/0/b/0b76e858431261de3412b2ef484233e5343e459b.png) **davinci-002** denies: ![image](https://us1.discourse-cdn.com/openai1/original/3X/b/b/bb492de3b034167da5fdbd807f8ed1f8d220dac3.png) ---
**Author:** ![treboralmasy1](https://avatars.discourse-cdn.com/v4/letter/t/ccd318/32.png) [@treboralmasy1](https://community.openai.com/u/treboralmasy1)\ **Post date:** [August 23, 2023, 3:21am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/7 "2023-08-23T03:21:36Z")
When I try to fine tune with gpt-3.5-turbo it is kicking me back an error: { “error”: { “message”: “Invalid base model: gpt-3.5-turbo-0613 (model must be one of ada, babbage, curie, davinci) or a fine-tuned model created by your organization: XXXXXX”, “type”: “invalid\_request\_error”, “param”: null, “code”: null } } Has anyone else experienced this or does anyone know a way to fix it? ---
**Author:** ![novaphil](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/novaphil/32/130934_2.png) [@novaphil](https://community.openai.com/u/novaphil)\ **Post date:** [August 23, 2023, 3:40am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/8 "2023-08-23T03:40:10Z")
> [@treboralmasy1](#): > > When I try to fine tune with gpt-3.5-turbo it is kicking me back an error: Are you using the new [FineTuningJob](https://platform.openai.com/docs/api-reference/fine-tuning/create) endpoint? ---
**Author:** ![treboralmasy1](https://avatars.discourse-cdn.com/v4/letter/t/ccd318/32.png) [@treboralmasy1](https://community.openai.com/u/treboralmasy1)\ **Post date:** [August 23, 2023, 3:52am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/9 "2023-08-23T03:52:30Z")
I’m actually using orhanerday (for php). Specifically with this: $result = $open\_ai-\>createFineTune([ “training\_file” =\> “file-XXX”, ‘model’=\>“gpt-3.5-turbo-0613”, ]); ---
**Author:** ![novaphil](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/novaphil/32/130934_2.png) [@novaphil](https://community.openai.com/u/novaphil)\ **Post date:** [August 23, 2023, 3:54am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/10 "2023-08-23T03:54:03Z")
I haven’t had a chance to test, but wouldn’t be surprised if you need to use the new API endpoint to fine-tune 3.5. So you’ll either need to write your own curl code or wait for the library to update. ---
**Author:** ![treboralmasy1](https://avatars.discourse-cdn.com/v4/letter/t/ccd318/32.png) [@treboralmasy1](https://community.openai.com/u/treboralmasy1)\ **Post date:** [August 23, 2023, 4:03am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/11 "2023-08-23T04:03:34Z")
Actually your answer put me on the correct path. For anyone needing help: in the Url.php file, replace: public static function fineTuneUrl(): string { return self::OPEN\_AI\_URL . “/fine-tunes”; } With: public static function fineTuneUrl(): string { return self::OPEN\_AI\_URL . “/fine\_tuning/jobs”; } ---
**Author:** ![Lawrence111](https://avatars.discourse-cdn.com/v4/letter/l/b5a626/32.png) [@Lawrence111](https://community.openai.com/u/Lawrence111)\ **Post date:** [August 24, 2023, 7:18am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/12 "2023-08-24T07:18:21Z")
I’m looking to fine-tune the `gpt-3.5-turbo` to mimic a specific character’s tone. What’s the best way to go about assembling the training set? At times, the character responds to another speaker’s queries, initiates the conversation topics, or simply engages in monologues. Given this, the roles of the `user` and `assistant` aren’t always perfectly aligned. Would it be acceptable if the `assistant`’s inputs outnumber those of the `user`? ---
**Author:** ![Lawrence111](https://avatars.discourse-cdn.com/v4/letter/l/b5a626/32.png) [@Lawrence111](https://community.openai.com/u/Lawrence111)\ **Post date:** [August 24, 2023, 7:27am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/13 "2023-08-24T07:27:43Z")
By the way, how to rename a fine-tuned model? ---
**Author:** ![\_j](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/_j/32/766292_2.png) [@\_j](https://community.openai.com/u/_j)\ **Post date:** [August 24, 2023, 9:46am UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/14 "2023-08-24T09:46:30Z")
gpt-3.5-turbo also can take a “name” parameter. This hasn’t been shown yet in fine-tune examples, but this is how it could go: ```auto { "messages": [ { "role": "system", "content": "You are Harry Potter. You chat with both Hermione Granger and Ron Weasley." }, { "role": "user", "name": "Hermione", "content": "What do you think of Ron Weasley, Harry?" }, { "role": "assistant", "name": "Harry", "content": "Well, Hermione, Ron is my best mate and a really loyal friend. Sure, he might have his moments of goofiness, but he's got a heart of gold. We've been through so much together, and I know I can always count on him. Plus, he's got a knack for lightening the mood, even in the darkest times. So yeah, I think Ron's a great guy and I'm lucky to have him by my side." } ] } ``` The AI doesn’t actually have a way to add a “name” to its own responses, this would only be seen in role inputs including past conversation. * * * Omitting either the “user” or “assistant” part of past conversation is often done, so that the chat history shows recent topics or responses without excessive contextual information. I would reconsider carefully using such a technique in _fine-tuning_ though, if it would not actually appear the exact same in AI chatbot use. Not verified on new endpoint: > [@Custom Fine-tune names](https://community.openai.com/t/custom-fine-tune-names/15787): > > Hi everyone! Quick update, we’ve added the ability to customize your fine-tune model names with a [suffix parameter](https://beta.openai.com/docs/guides/fine-tuning/customize-your-model-name). openai api fine\_tunes.create -t test.jsonl -m ada --suffix "custom model name" Would create a model named: ada:ft-your-org:custom-model-name-2022-02-15-04-21-04 Have a great weekend! (PS remember to upgrade first: pip install --upgrade openai) ---
**Author:** ![brightj](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/brightj/32/11130_2.png) [@brightj](https://community.openai.com/u/brightj)\ **Post date:** [August 25, 2023, 6:10pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/15 "2023-08-25T18:10:29Z")
Can a fine-tuned model be used by other people with their own API keys? Like if I write open-source software based around one and try to share it, will nobody else be able to use it? ---
**Author:** ![novaphil](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/novaphil/32/130934_2.png) [@novaphil](https://community.openai.com/u/novaphil)\ **Post date:** [August 25, 2023, 6:11pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/16 "2023-08-25T18:11:57Z")
Only if you add them to your Organization, which doesn’t sound like a great idea for this model. There’s no way to make a fine-tuned model public or to share it. Best you can do is share the training file and instructions on how to have users create their own fine-tuned model on their account. ---
**Author:** ![ryanmelo87](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ryanmelo87/32/178996_2.png) [@ryanmelo87](https://community.openai.com/u/ryanmelo87)\ **Post date:** [September 1, 2023, 9:47pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/17 "2023-09-01T21:47:20Z")
I’d love to add these features (Website Crawlers to read the website full information + file uploaded with data parser) then integrated with OpenAI with 3.5 fine-tune, is this idea achieveable? ---
**Author:** ![Nexoid](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/nexoid/32/195641_2.png) [@Nexoid](https://community.openai.com/u/Nexoid)\ **Post date:** [September 9, 2023, 3:37pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/18 "2023-09-09T15:37:50Z")
I read the documentation last week and they said it was “coming” ---
**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 15, 2023, 12:29pm UTC](https://community.openai.com/t/gpt-3-5-turbo-fine-tuning-now-available-and-new-gpt3-models/327425/19 "2023-12-15T12:29:43Z")