@browser_use: Nemotron 3.5 Super VL is good for browser use but mogged by Luna > 48.5/100 vs GPT-6 Luna’s 67.7/100 > at 61% higher es…
Summary
Nemotron 3.5 Super VL performs well on browser-use tasks but trails GPT-6 Luna, scoring 48.5/100 versus 67.7/100 at an estimated 61% higher model cost — solid for a newcomer.
View Cached Full Text
Cached at: 10/17/26, 12:47 AM
Nemotron 3.5 Super VL is good for browser use but mogged by Luna 👀
> 48.5/100 vs GPT-6 Luna’s 67.7/100 > at 61% higher estimated model cost
Not bad for a newcomer. https://t.co/qNQlSCJo8p
Similar Articles
@browser_use: Browser Use Bench v2 Pareto frontier got completely redrawn today > Claude Opus 5.5: 59.4 > GPT‑6 Sol medium: 66.9 (3.5…
The Browser Use Bench v2 benchmark has updated its Pareto frontier, showcasing GPT-6 models from OpenAI outperforming and being more cost-effective than Claude Opus 5.5 from Anthropic.
@browser_use: Astra is a monster at browser use
Astra achieved 77.3% on the Browser Use Benchmark v2, far surpassing Opus 5 (50.5%) and GPT-5.6 Sol xhigh (49.1%), with 22 of 60 tasks earning full marks compared to zero for Opus 5.
@browser_use: Opus 5 and GPT-5.6 Sol are neck-and-neck on this!
Alexander Yue introduces a new browser-use benchmark where Opus 5 and GPT-5.6 Sol show similar performance, emphasizing the benchmark's robust design with verified rubrics for LLM judges.
@reach_vb: GPT-5.6 Luna Max scores 13.3 points above Sonnet 5 Max on DeepSWE v1.1 while Sonnet costs 44x as much. DeepSWE tests co…
GPT-5.6 Luna Max outperforms Sonnet 5 Max on the DeepSWE v1.1 coding benchmark at a much lower cost, and also shows strong performance compared to Gemini 3.7 Flash Medium.
@browser_use: Try out our new agent with GPT 5.6 Luna
browser-use announces a new agent powered by GPT-5.6 Luna, which can read top Hacker News posts and generate a full report for only 3 cents, highlighting the low cost of AI-driven web browsing.