12 ms·

If GPT 5 truly has 400k context, that might be all it needs to meaningfully surpass Opus.

AS04 1y ago
400k context with 100% on the fiction livebench would make GPT-5 the undisputably best model IMHO. Don't think it will achieve that though, sadly.
simonw 1y ago
It's 272,000 input tokens and 128,000 output tokens.
- zurfer 1y ago
  Woah that's really kind of hidden. But I think you can specify max output tokens. Need to test that!
- 6thbit 1y ago
  Oh, I had not grasped that the “context window” size advertised had to include both input and output. But is it really 272k even if the output was say 10k? Cause it does say “max output” in the docs, so I wonder
  - simonw 1y ago
    This is the only model where the input limit and the context limit are different values. OpenAI docs team are working on updating that page.
- dudeinhawaii 1y ago
  The website clearly lays them out as 400k input and 128k output [1]. I just updated my AI apps to support the new models. I routinely fill the entire context on large code calls. Input is not a "shared" context. I found 100k was barely enough for a single project without spillover, so 4x allows for linking more adjacent codebases for large scale analysis. [1] https://platform.openai.com/docs/models/gpt-5 https://platform.openai.com/docs/models/gpt-5
dimal 1y ago
Even with large contexts there's diminishing returns. Just having the ability to stuff more tokens in context doesn't mean the model can effectively use it. As far as I can tell, they always reach a point in which more information makes things worse.
Byamarro 1y ago
More of a question is its context rot tendency than the size of its context :) LLMs are supposed to load 3 bibles into their context, but they forget what they were about to do after loading a 600LoC of locales.
andrewmutz 1y ago
Having a large context window is very different from being able to effectively use a lot of context. To get great results, it's still very important to manage context well. It doesn't matter if the model allows a very large context window, you can't just throw in the kitchen sink and expect good results
tekacs 1y ago
Coupled with the humungous price difference...

AS04 1y ago

400k context with 100% on the fiction livebench would make GPT-5 the undisputably best model IMHO. Don't think it will achieve that though, sadly.

simonw 1y ago

It's 272,000 input tokens and 128,000 output tokens.

zurfer 1y ago
Woah that's really kind of hidden. But I think you can specify max output tokens. Need to test that!
6thbit 1y ago
Oh, I had not grasped that the “context window” size advertised had to include both input and output. But is it really 272k even if the output was say 10k? Cause it does say “max output” in the docs, so I wonder
- simonw 1y ago
  This is the only model where the input limit and the context limit are different values. OpenAI docs team are working on updating that page.
dudeinhawaii 1y ago
The website clearly lays them out as 400k input and 128k output [1]. I just updated my AI apps to support the new models. I routinely fill the entire context on large code calls. Input is not a "shared" context. I found 100k was barely enough for a single project without spillover, so 4x allows for linking more adjacent codebases for large scale analysis. [1] https://platform.openai.com/docs/models/gpt-5 https://platform.openai.com/docs/models/gpt-5

dimal 1y ago

Even with large contexts there's diminishing returns. Just having the ability to stuff more tokens in context doesn't mean the model can effectively use it. As far as I can tell, they always reach a point in which more information makes things worse.

Byamarro 1y ago

More of a question is its context rot tendency than the size of its context :) LLMs are supposed to load 3 bibles into their context, but they forget what they were about to do after loading a 600LoC of locales.

andrewmutz 1y ago

Having a large context window is very different from being able to effectively use a lot of context. To get great results, it's still very important to manage context well. It doesn't matter if the model allows a very large context window, you can't just throw in the kitchen sink and expect good results

tekacs 1y ago

Coupled with the humungous price difference...