LMSYS Chatbot Arena, every mention

21 scenes · ← back to LMSYS Chatbot Arena

tap a year for its mentions
0040280420252026episodesmentions
02420252026episodes it came up in
0010220420252026episodesmentions per episode

every year anyone Anjney Midha 55Wei-Lin Chiang 6Oliver Wang 3Ion Stoica 2Simon Moe 1Dylan Patel 1

Verbatim, from the transcripts: the passages where LMSYS Chatbot Arena comes up

loading…

How Open Source Became AI's Backbone | Inferact with a16z Aug 5, 2026 · 1 mention

  • ▶ 39:39 Simon Moe One of the very useful, uh, benchmarks that we have on Arena for, uh, for case three has been front-end coding.

Where does consumer AI stand at the end of 2025? Dec 29, 2025 · 1 mention

  • ▶ 26:05 unnamed speaker Let's top the Elemarina charts.

Google DeepMind Developers: How Nano Banana Was Made Oct 28, 2025 · 3 mentions

  • ▶ 2:21 Oliver Wang I, so I actually didn't feel like it was going to go viral until we had released on Ella Marina. 3 times in the scene

Dylan Patel on GPT-5’s Router Moment, GPUs vs TPUs, Monetization Aug 18, 2025 · 1 mention

  • ▶ 1:03:40 Dylan Patel Like they're on LLM arena right now and they're pretty decent there, but it's like, that's just like a sycophantic model.

Beyond Leaderboards: LMArena’s Mission to Make AI Reliable May 29, 2025 · 63 mentions

  • ▶ 0:37 Anjney Midha One of the arenas misunderstood partly because we're just early in AI. 6 times in the scene
  • ▶ 3:52 Anjney Midha You know, at this point, Arena is hard to miss in the AI space, you know, whether it's Grok three coming out and Elon putting it up for the bulk of the keynote or, you know, Demis, uh, using the web dev arena scores to 5 times in the scene
  • ▶ 6:27 Anjney Midha And for two reasons, I find, I've always found LM Arena interesting. 3 times in the scene
  • ▶ 14:27 Anjney Midha It was insufficient to keep measuring the progress of the models on coding with the base design of the platform, which is just Chatbot Arena. 2 times in the scene
  • ▶ 19:19 Wei-Lin Chiang That was the very beginning of why we believe arena is fundamentally different. 5 times in the scene
  • ▶ 24:07 Anjney Midha So it's in reality for critics who say, you know, these exam, static exams are hard tests and Arena is an easy to gain benchmark. 5 times in the scene
  • ▶ 29:55 Anjney Midha LM Arena started as a research project. 5 times in the scene
  • ▶ 39:21 Anjney Midha In many ways, I feel like this is a, the birth of ARENA couldn't have happened anywhere else other than an interdisciplinary lab at a university research, fundamental research university like Berkeley.
  • ▶ 46:47 Ion Stoica When Shellbot Arena started, it was a lot of excitement and so forth, but then it was, you think, okay, uh, it was a feeling, at least for some people in the group, that we are done here. 2 times in the scene
  • ▶ 48:35 Anjney Midha I think there's a, a chart I saw recently that showed that, that compared the number of models being released and tested on Ella Marina per year over the last two years. 6 times in the scene
  • ▶ 1:12:21 Anjney Midha We spent a bunch of time talking about how ARENA was born and how the, the, the big idea, at least the theoretical idea is that for, to unlock more reliability in AI, we need more testing of AI. 5 times in the scene
  • ▶ 1:20:07 Anjney Midha And yet on Arena, 7 times in the scene
  • ▶ 1:28:19 Anjney Midha One of the things that surprised me when I looked at the repo, you know, it's an open source repo, is how well the model performed on Arena. 5 times in the scene
  • ▶ 1:31:43 Anjney Midha The arena experience will look different than ChatGPT. 2 times in the scene
  • ▶ 1:34:44 Anjney Midha I think from day one, LM Arena has open sourced
  • ▶ 1:39:44 Wei-Lin Chiang Yeah, so for example, this real-world testing idea of Arena can be applied to many different applications like we discussed, right?
  • ▶ 1:43:38 Anjney Midha In some sense, it seems like you guys were actually a little bit ahead of the curve where Arena has always been an environment for agents more than, you know, a set of static evals. 2 times in the scene
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 1,000 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.