# Nsight Copilot

**URL:** https://forums.developer.nvidia.com/c/developer-tools/nsight-copilot/739.md

[Latest](https://forums.developer.nvidia.com/latest.md) · [Categories](https://forums.developer.nvidia.com/categories.md) · [Tags](https://forums.developer.nvidia.com/tags.md)

---

## [Nsight-copilot cuda-docs MCP endpoint returns HTTP 500 on every query (internal RAG URL leaked in error)](https://forums.developer.nvidia.com/t/nsight-copilot-cuda-docs-mcp-endpoint-returns-http-500-on-every-query-internal-rag-url-leaked-in-error/380756)

<div class="topic-metadata">

**Author:** [@tnatkinn](https://forums.developer.nvidia.com/u/tnatkinn)\
**Replies:** 2\
**Last updated:** [August 21, 2026, 6:09am UTC](https://forums.developer.nvidia.com/t/nsight-copilot-cuda-docs-mcp-endpoint-returns-http-500-on-every-query-internal-rag-url-leaked-in-error/380756 "2026-08-21T06:09:43Z")

</div>

The hosted CUDA-docs MCP endpoint connects and authenticates fine, but every single tool call fails with an HTTP 500 originating from an internal service. Endpoint https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-doc…

---

## [Request for NIM API Rate Limit Quota Increase – AI Agent Development](https://forums.developer.nvidia.com/t/request-for-nim-api-rate-limit-quota-increase-ai-agent-development/373429)

<div class="topic-metadata">

**Author:** [@johannschmidt829](https://forums.developer.nvidia.com/u/johannschmidt829)\
**Replies:** 0\
**Last updated:** [June 15, 2026, 10:17pm UTC](https://forums.developer.nvidia.com/t/request-for-nim-api-rate-limit-quota-increase-ai-agent-development/373429 "2026-06-15T22:17:28Z")

</div>

I am writing to request a rate limit increase for my NVIDIA NIM API developer account. Account Details: Registered Email: johannschmidt829@gmail.com Current tier: 40 RPM Requested Limit: 200 RPM Use Case: I am devel…

---

## [Default Debug Debug Terminal Keybinding](https://forums.developer.nvidia.com/t/default-debug-debug-terminal-keybinding/364711)

<div class="topic-metadata">

**Author:** [@frenzi](https://forums.developer.nvidia.com/u/frenzi)\
**Replies:** 0\
**Last updated:** [March 25, 2026, 10:04am UTC](https://forums.developer.nvidia.com/t/default-debug-debug-terminal-keybinding/364711 "2026-03-25T10:04:39Z")

</div>

Hi, I don’t think Ctrl+Shift+R is a particularly good choice for the default keybinding for Nsight Coplilot: Debug Terminalsince this is also the default binding for the Refactor action in VSCode. Of course it only takes…

---

## [CRITICAL ISSUE: Nsight Copilot Providing FACTUALLY INCORRECT CUDA Knowledge on NVFP4💥](https://forums.developer.nvidia.com/t/critical-issue-nsight-copilot-providing-factually-incorrect-cuda-knowledge-on-nvfp4/352854)

<div class="topic-metadata">

**Author:** [@913363043](https://forums.developer.nvidia.com/u/913363043)\
**Replies:** 1\
**Last updated:** [December 8, 2025, 1:32am UTC](https://forums.developer.nvidia.com/t/critical-issue-nsight-copilot-providing-factually-incorrect-cuda-knowledge-on-nvfp4/352854 "2025-12-08T01:32:59Z")

</div>

Dear Team, I am raising a major concern regarding the quality and accuracy of the code and architectural knowledge provided by Nsight Copilot, which is supposedly built on a robust “CUDA Knowledge Model.” I requested a…

---

## [Likely cause: not authorized to access the model deployment](https://forums.developer.nvidia.com/t/likely-cause-not-authorized-to-access-the-model-deployment/350905)

<div class="topic-metadata">

**Author:** [@1033287073](https://forums.developer.nvidia.com/u/1033287073)\
**Replies:** 3\
**Last updated:** [November 28, 2025, 5:18am UTC](https://forums.developer.nvidia.com/t/likely-cause-not-authorized-to-access-the-model-deployment/350905 "2025-11-28T05:18:06Z")

</div>

403 ERROR Likely cause: not authorized to access the model deployment. Why did this happen?

---

## [Nsight Copilot is down?](https://forums.developer.nvidia.com/t/nsight-copilot-is-down/350662)

<div class="topic-metadata">

**Author:** [@vinayrjumani](https://forums.developer.nvidia.com/u/vinayrjumani)\
**Replies:** 1\
**Last updated:** [November 10, 2025, 8:01pm UTC](https://forums.developer.nvidia.com/t/nsight-copilot-is-down/350662 "2025-11-10T20:01:24Z")

</div>

I have been using Nsight copilot for almost a week, it worked pretty well, gave me the kernels that I requested and was able to understand all kernels. Now, it doesn’t seem to work. Every time I prompt something, it cras…

---

## [Weird answer from nsight copilot](https://forums.developer.nvidia.com/t/weird-answer-from-nsight-copilot/350533)

<div class="topic-metadata">

**Author:** [@ingridli1998](https://forums.developer.nvidia.com/u/ingridli1998)\
**Replies:** 4\
**Last updated:** [November 7, 2025, 5:25pm UTC](https://forums.developer.nvidia.com/t/weird-answer-from-nsight-copilot/350533 "2025-11-07T17:25:35Z")

</div>

I think this answer is very ridiculous.

---

## [Nsight Copilot Issues](https://forums.developer.nvidia.com/t/nsight-copilot-issues/347107)

<div class="topic-metadata">

**Author:** [@guilhermehartmann](https://forums.developer.nvidia.com/u/guilhermehartmann)\
**Replies:** 2\
**Last updated:** [October 15, 2025, 1:00pm UTC](https://forums.developer.nvidia.com/t/nsight-copilot-issues/347107 "2025-10-15T13:00:26Z")

</div>

I am having some strange error after login into the Nsight Copilot, first I have “no chat model selected” but looking further I can see this error on the extension : \[error\] TypeError: prompts is not iterable at Prompt…

---

## [How to enable CUDA aware chat and CUDA code generation windows in VS code?](https://forums.developer.nvidia.com/t/how-to-enable-cuda-aware-chat-and-cuda-code-generation-windows-in-vs-code/343200)

<div class="topic-metadata">

**Author:** [@orong13](https://forums.developer.nvidia.com/u/orong13)\
**Replies:** 5\
**Last updated:** [October 1, 2025, 4:28pm UTC](https://forums.developer.nvidia.com/t/how-to-enable-cuda-aware-chat-and-cuda-code-generation-windows-in-vs-code/343200 "2025-10-01T16:28:59Z")

</div>

After installing the Nsight Copilot extension and signing up using my account, what shall I do in order to enable the chat to start using the copilot?

---

## [Context not working but seems to be expected](https://forums.developer.nvidia.com/t/context-not-working-but-seems-to-be-expected/342962)

<div class="topic-metadata">

**Author:** [@federicocusot](https://forums.developer.nvidia.com/u/federicocusot)\
**Replies:** 2\
**Last updated:** [September 11, 2025, 4:34pm UTC](https://forums.developer.nvidia.com/t/context-not-working-but-seems-to-be-expected/342962 "2025-09-11T16:34:08Z")

</div>

Seem like the model can’t get access to the file even though I provide the file as context on the the chat box.
