author:maxbittker

· 18 artifacts, sorted by favorites. · open in search — combine tags, sort, filter by date →

@maxbittker 2026-08-12 — Grok 4.6 is Runescape bench's new #1, at only ~60% the cost of previous leader Fable 5. https://t.co/Sfeb5QJo3T ♥1328
@maxbittker 2026-06-16 — GLM-5.2 just scored better than Opus 4.7 and GPT 5.4 on Runescape bench. These models were best in class only 2-3 mont ♥729
@maxbittker 2026-10-05 — The code Astra and Opus 5.5 write when they're activating their RL super-powers (animation and modeling) is so freaky an ♥390
@maxbittker 2026-10-13 — Runescape Bench result - Gemini 4 Argon is legit! Very similar to Astra in performance, at a Sol price. https://t.co/Zw ♥301
@maxbittker 2026-03-11 — Gemini 3 Flash is a crazy standout in Runescape benchmarking. This is flash fishing 3 different setups (shrimp->salmon- ♥192
@maxbittker 2026-10-07 — on Runescape Bench, Haiku 5.5 scores above Opus 4.6, released 8 months ago, for 1/20 the price. Haiku 5.5 is outclassed ♥129
@maxbittker 2026-07-10 — New results on the Runescape benchmark: - GPT-5.6 Sol tied with Fable - Grok 4.5 scored above Opus 4.8 for 1/2 the price ♥128
@maxbittker 2026-06-30 — Sonnet 5's Runescape benchmark scores are big improvement over Sonnet 4.6's, but outclassed by Pareto frontier neighbors ♥98
@maxbittker 2026-10-12 — Muse Wonder does.... fine on Runescape bench, scores like GPT-6 Luna at 35x price. Looking at the runs, I will say tha ♥93
@maxbittker 2026-07-17 — Kimi K3 did poorly on Runescape bench because it thinks a lot and has slow inference. Performs worse than Kimi K2.7, way ♥70
@maxbittker 2026-08-12 — Grok 4.6 makes a ton of tiny tool calls, which means the cost ends up being similar to terra and opus. Turns out to be ♥59
@maxbittker 2026-09-22 — Just ran Deepseek v4.1 Flash on Runescape bench as well. Such a remarkable Pareto outlier, been a huge boon at @websim a ♥37
@maxbittker 2026-09-28 — In a multi-agent evaluation on ProgramBench: Sonnet 5.5's fastest setup was subagents Opus 5.5 went fastest using a 5 a ♥28
@maxbittker 2026-06-17 — @thaonlyjonathan $150 at api price- Fable was cheaper than GPT-5.5 xhigh. And for reference, GLM-5.2 was $32 https://t. ♥18
@maxbittker 2026-05-19 — feeling the reinforcement learning... Gemini 3.5 Flash is tied with GPT-5.5 at navigating complex tasks in Runescape - ♥18
@maxbittker 2026-07-10 — GPT 5.6 Sol, Luna, and Grok 4.5 scores on Runescape bench individual skills. Scoring well on the skills near the right h ♥14
@maxbittker 2026-08-21 — @leo_linsky gemini flash 3.7, grok 4.6, luna, Deepseek flash v4 Roughly $100 per 60m trial for 22 agents ♥13
@maxbittker 2026-09-04 — @asaridarr "bad" skills: strength (attempted waterfall quest but died) and woodcutting (fine strategy but didn't find as ♥9