Skip to content
Problems.

API rate limits and usage caps block real work

LLM apps16 posts from 16 people12d active+33 posts in the last 7 days, 3 the 7 days before (steady)

Posts per day

Posts per day16 posts, Aug 28 to Sep 25

The posts behind it

16, newest first
PostDate
Qwen 3.8 27b be like...The user is frustrated - I rambled too much and didn't act. Let's just run the test suite and move on. No more forensics. One command, execute, then report.. (Original memo is a casual internal monologue in English. Translating faithfully while preserving the informal,...r/LocalLLaMAu/Graemer71Sep 24yesterday
core: docstring examples use nonexistent imports, retired Claude models and invalid `openai:` model strings### Submission checklist - [x] This is a bug, not a usage question. - [x] I added a clear and descriptive title that summarizes this issue. - [x] I used the GitHub search to find a similar question and didn't find it. - [x] I am sure that this is a bug in LangChain rather than...langchain-ai/langchainaman99dexSep 24yesterday
...rate limits, etc.) and points at the vendor's own API docs for anything else. The annoying part isn't tool calling, it's auth. You don't want the agent managing OAuth tokens for 10+ apps. A passthrough proxy that holds the credentials, injects them per request, and scopes...Hacker News commentsbyungkyuSep 223 days ago
[source-facebook-marketing] Ad-account quota errors (code 17/2446079) retried as transient, causing repeated full sync failures...lasts (minutes to an hour). Net effect: the connector spends its entire retry budget failing fast inside a block window it never even detects, then gives up and forces a full Airbyte-level sync restart - which just re-enters the same block. ### Relevant log output Observed on...airbytehq/airbyteFVCCarneiroSep 1510 days ago
...indistinguishable from legit users, but en masse can take down a VPS. Rate limiting doesn’t work when individual IPs are only making a handful of requests. So that’s why we’re using CF/Fastly. Too bad, I’d prefer not to, but there is just not an alternative at this point.Hacker News commentskarlsheaSep 1411 days ago
Designing a local AI harness against the "echo-chamber" loop: deterministic arbiters, tight token caps, and skeptical devs...use RTK (rust token killer) as an opencode plugin. . Hard Circuit Breaker. If the agent fails to produce code that passes mise run test within 3 attempts, the harness halts the loop.. My questions are : . * Deterministic oracles and models feedback : when using binary (exit 0/1)...r/devopsu/Brilliant_Willow1311Sep 1312 days ago
ollama-cloud: expose per-model metadata in the model API mirroring other providers## Summary `/api/show` and `/v1/models` currently expose only a boolean `thinking` capability and a `details` block (family, param size, quantization). They do **not** report, per model: - supported `reasoning_effort` levels, - maximum output-token limit, - context window (only...ollama/ollamafgrehmSep 112 weeks ago
...model you might be better off directly using their API, which is unfortunately far more painful than using OpenRouter, but it might at least be more reliable for your product. I think the rate limit issue happens because of OpenRouter sending so many requests to Google.Hacker News commentsjwxzSep 112 weeks ago
Grok Bot Ultra users: how do you make the weekly allowance last? Mine reached 99% in three daysHi everyone, I’ve used Cursor for years, and I’m really excited about Grok Bot. It’s the kind of tool I want to build my daily workflow around. But my first experience with Ultra’s usage limits has left me frustrated an…Cursor ForumolegdaterSep 102 weeks ago
Provider errors returned in a 200 response body are never classified...fault with **HTTP 200 whose body carries `{"error": {"code": 502, ...}}`** rather than a failing status. This is documented gateway behaviour, not a malformed response. * I expect the same exception I'd get if that status had arrived on the response line - for a 502, something...langchain-ai/langchainyannsaySep 102 weeks ago
Allow to use SSH instead of HTTPS to fetch modules from terraform registry### Terraform Version ### Use Cases When using a module from the terraform registry, e.g. `module "some_bucket" { source = "terraform-aws-modules/s3-bucket/aws" ... } ` I would like to be able to tell terraform to use an SSH github url to checkout this module. It seems that...hashicorp/terraformrikostave1234Sep 43 weeks ago
...+ design discussion, Opus 5 agents for implementation" paradigm. But an incredibly annoying UX problem is that the resulting report from the agents that Fable reads isn't surfaced to us in the main dialog, it's only summarized back to us (unless you idle in the agent's window to...Hacker News commentsashkankianiSep 23 weeks ago
...of the UI rather than the actual speed of user interactions. Also the menu bar icon doesn't work for me (on MacOs). I can see it, but it doesn't do anything and there's nothing in the app itself to enable/disable it (though thankfully it does respect cmd+drag to delete).Hacker News commentssceptic123Sep 13 weeks ago
structuredContent is dropped from isError=True tool results (but preserved on success)## Summary `_convert_call_tool_result` extracts `CallToolResult.structuredContent` into the `MCPToolArtifact` only on the success path. When `isError=True` it raises `_MCPToolExecutionError` **before** `structuredContent` is read, so a machine-readable error payload is silently...langchain-ai/langchainglenlouis8Aug 293 weeks ago
...keeps on switching providers. I'm paying for every token... why rate limit me? It was unusable compared to just using the official Deepseek provider which has a much better cache rate.Hacker News commentsDefineOutsideAug 284 weeks ago
[source-paypal-transaction] transactions stream exhausts retries on PayPal 429 rate limits due short default backoff...happened? During the sync ### Relevant information The PayPal Transaction source is failing consistently while syncing the `transactions` stream. The `balances` stream completes successfully, but `transactions` enters `RATE_LIMITED` and repeatedly fails with PayPal HTTP 429...airbytehq/airbytejaska120Aug 284 weeks ago

Companies and products named

Company or productPosts naming it
DeepSeek4
Ollama3
LangChain3
Model Context Protocol2
OpenRouter2
Anthropic2
About this problem

Evidence

16 posts from 16 people in 8 places, about 4 a week over 28 days. Mostly on Hacker News comments, langchain-ai/langchain, airbytehq/airbyte. Tools named alongside: DeepSeek, Ollama, LangChain, Model Context Protocol.

Frustration Frustration 1 of 3· Seen on GitHub issues, Hacker News, Reddit, Community forums

How it was grouped

Posts that state a pain and match the "rate-limits" rule. First post Aug 28, 2026, latest Sep 24, 2026. Corroborated: 3 or more posts from 2 or more people or places. Method

History

  • 2026-09-25 Momentum: rising to steady
  • 2026-09-25 Added: API rate limits and usage caps block real work (13 posts)

Rising problems by email

Mondays: the problems in data, tech and AI that grew fastest that week.

Double opt-in. Unsubscribe any time.