---
format: "aidr-story-markdown/v1"
id: "61020cdda2cdcf26cc62f2dcabb69e84d856e492f43ca7612c29ccea49f8ddcb"
canonical_url: "https://aidr.today/61020cdd?lang=en"
title: "SGLang AI Beats Level 9 Fox in Super Smash Bros Melee"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-01T01:07:38.000Z"
category: "Agents"
topics: ["agent"]
source_urls: ["https://huggingnews.com/ai/update-sglang-ai-beats-level-9-fox-in-super-smash-bros-melee-187c1c9a","https://marketbrief.now/ai/update-sglang-ai-beats-level-9-fox-in-super-smash-bros-melee-187c1c9a","https://huggingnews.com/ai/tavus-ai-fools-48percent-of-humans-in-first-video-turing-test-pass-170f5fb1","https://marketbrief.now/ai/tavus-ai-fools-48percent-of-humans-in-first-video-turing-test-pass-170f5fb1","https://the-decoder.com/nearly-half-of-test-subjects-mistook-tavus-ai-video-avatar-for-a-real-person-on-a-one-minute-call/"]
summary: "← Back to live feed · 1 stories across 1 day Recent testing of multimodal agents has shown an ability to process and react to live visual data in high-speed gaming environments. An agent using the SGLang framework defeated a Level 9 computer opponent in Super Smash Bros Melee, playing as the character Fox on the Final Destination map without items. The victory, which included a 2 stock of the CPU player, was executed in real time through an API. This development builds on a previous success where a Qwen3.8-27B model beat the Pokémon FireRed champion with responses in under 100 ms. SGLang utilizes a /v1/decisions endpoint to turn large models into classification and scoring tools, alongside a /v1/systemone endpoint for TypeSafe SDK integration. Some observers of earlier tests noted that the majority of the model's critical tactical choices were poor, suggesting the results are primarily a achievement in systems integration."
---

# SGLang AI Beats Level 9 Fox in Super Smash Bros Melee

> [Open the canonical story](<https://aidr.today/61020cdd?lang=en>)

**Published:** 2026-10-01T01:07:38.000Z
**Category:** Agents
**Topics:** agent

## Summary

← Back to live feed · 1 stories across 1 day Recent testing of multimodal agents has shown an ability to process and react to live visual data in high\-speed gaming environments\. An agent using the SGLang framework defeated a Level 9 computer opponent in Super Smash Bros Melee, playing as the character Fox on the Final Destination map without items\. The victory, which included a 2 stock of the CPU player, was executed in real time through an API\. This development builds on a previous success where a Qwen3\.8\-27B model beat the Pokémon FireRed champion with responses in under 100 ms\. SGLang utilizes a /v1/decisions endpoint to turn large models into classification and scoring tools, alongside a /v1/systemone endpoint for TypeSafe SDK integration\. Some observers of earlier tests noted that the majority of the model's critical tactical choices were poor, suggesting the results are primarily a achievement in systems integration\.

## Sources

- [Story source](<https://huggingnews.com/ai/update-sglang-ai-beats-level-9-fox-in-super-smash-bros-melee-187c1c9a>)
- [Story source](<https://marketbrief.now/ai/update-sglang-ai-beats-level-9-fox-in-super-smash-bros-melee-187c1c9a>)
- [Story source](<https://huggingnews.com/ai/tavus-ai-fools-48percent-of-humans-in-first-video-turing-test-pass-170f5fb1>)
- [Story source](<https://marketbrief.now/ai/tavus-ai-fools-48percent-of-humans-in-first-video-turing-test-pass-170f5fb1>)
- [Story source](<https://the-decoder.com/nearly-half-of-test-subjects-mistook-tavus-ai-video-avatar-for-a-real-person-on-a-one-minute-call/>)