# Problem for large context window (400k)

**URL:** <https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864>\
**Category:** Models\
**Created:** [April 15, 2024, 2:03pm UTC](https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864 "2024-04-15T14:03:53Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Trillia](https://avatars.discourse-cdn.com/v4/letter/t/e9c0ed/32.png) [@Trillia](https://discuss.huggingface.co/u/Trillia)\
**Post date:** [April 15, 2024, 2:03pm UTC](https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864/1 "2024-04-15T14:03:53Z")

</div>

Hi,

I’d like to use the model namespace-Pt/activation-beacon-llama2-7b-chat. It’s a model with a 400k context window, which is what I want.  
But when I try to launch a dedicated Inference Endpoints, when I use it, it says {‘error’: 'Input validation error: `inputs` tokens + `max_new_tokens` must be \<= 2048. […] }.  
I already tried to modify Max Input Length (per Query) and Max Number of Tokens (per Query), but I have the same problem.

So how can I use this large context window model please ?

Thank you !

---

<div class="post-metadata">

**Author:** ![Rayisy](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/rayisy/32/26497_2.png) [@Rayisy](https://discuss.huggingface.co/u/Rayisy)\
**Post date:** [May 2, 2024, 11:38am UTC](https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864/2 "2024-05-02T11:38:21Z")

</div>

I need help!!!

---

<div class="post-metadata">

**Author:** ![storz](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/storz/32/28678_2.png) [@storz](https://discuss.huggingface.co/u/storz)\
**Post date:** [June 27, 2024, 2:59pm UTC](https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864/3 "2024-06-27T14:59:19Z")

</div>

No help, but just to let you know: I have a similar problem with the gradient llama3 models, I need a big context window, but the inference endpoint somehow does not play along nicely. Asked support, no answer yet.

---

<div class="post-metadata">

**Author:** ![GooglyEyeSuperman](https://avatars.discourse-cdn.com/v4/letter/g/67e7ee/32.png) [@GooglyEyeSuperman](https://discuss.huggingface.co/u/GooglyEyeSuperman)\
**Post date:** [July 2, 2024, 5:28pm UTC](https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864/4 "2024-07-02T17:28:41Z")

</div>

Hi also want to know!

---

<div class="post-metadata">

**Author:** ![alsterg](https://avatars.discourse-cdn.com/v4/letter/a/d78d45/32.png) [@alsterg](https://discuss.huggingface.co/u/alsterg)\
**Post date:** [July 24, 2024, 8:20am UTC](https://discuss.huggingface.co/t/problem-for-large-context-window-400k/81864/5 "2024-07-24T08:20:20Z")

</div>

Same here. I’m chasing my tail trying to deploy a large context window model on an inference endpoint.  
Some general guidance on how to do it would be nice.