News for hackers. Stuff that matters.
Re:GLM-5.2 is the new leading open weights model on Artifici... by CuriouslyC on Wednesday June 17, 2026 @06:41AM (#48568406)
Be careful about unofficial providers, a lot of them misconfigure models or stealth quantize them. For a while the difference between Kimi on the official API and most third party providers was 20-40%.
Re:GLM-5.2 is the new leading open weights model on Artifici... by cedws on Wednesday June 17, 2026 @06:59AM (#48568575)
OpenRouter should be penalising or banning for this.
Re:GLM-5.2 is the new leading open weights model on Artifici... by kilroy123 on Wednesday June 17, 2026 @07:56AM (#48569116)
This is my biggest complaint about OpenRouter and I'm a fan. Might be pretty tough at scale?
Re:GLM-5.2 is the new leading open weights model on Artifici... by alecco on Wednesday June 17, 2026 @08:27AM (#48569446)
Would that align with their VC-backed incentives?
Re:GLM-5.2 is the new leading open weights model on Artifici... by mrngld on Wednesday June 17, 2026 @02:34PM (#48574687)
If your users can't trust your product then I'd say that'd be a pretty strong incentive?
Re:GLM-5.2 is the new leading open weights model on Artifici... by orbital-decay on Wednesday June 17, 2026 @09:28AM (#48570276)
They have an "exacto" category with providers they supposedly verified
Re:GLM-5.2 is the new leading open weights model on Artifici... by ComputerGuru on Wednesday June 17, 2026 @10:23AM (#48570979)
That’s only for tool use.
Re:GLM-5.2 is the new leading open weights model on Artifici... by rsanek on Friday June 19, 2026 @05:06PM (#48603246)
That's just one part of it. According to https://openrouter.ai/docs/guides/routing/model-variants/exa...

  We use three classes of signals:
   * Tool-calling success and reliability from real traffic
   * Provider performance metrics such as throughput and latency
   * Benchmark and evaluation data as it becomes available
Re:GLM-5.2 is the new leading open weights model on Artifici... by unrvl22 on Wednesday June 17, 2026 @06:43AM (#48568423)
the 2 I mentioned both have a fairly large following, who run benchmarks and absolutely will spot issues.
Re:GLM-5.2 is the new leading open weights model on Artifici... by thehamkercat on Wednesday June 17, 2026 @08:36AM (#48569557)
Kimi K2 had a vendor verifier: https://github.com/MoonshotAI/K2-Vendor-Verifier

(there's a table which shows comparison between vendors)

Also, it seems there's a general one as well (for all kimi models?): https://github.com/MoonshotAI/Kimi-Vendor-Verifier