Grok 4.5 Shows Strong Coding Efficiency
XFreeze · x · 2026-07-15
According to a repost, Grok 4.5 demonstrates exceptional coding efficiency on the FrontierSWE benchmark: achieving top scores while using significantly fewer tokens.
Comparisons provided:
- Against Fable 5, average total tokens are 20.1M vs 116.9M
- Average wall-clock time is 229 minutes vs 450 minutes
- Token usage is nearly 6x lower, and time is cut in half
The core takeaway is that Grok 4.5 achieves high scores through resource efficiency rather than brute-forcing with massive token counts and compute.
Related event: Grok 4.5 Rises to No. 2 on FrontierSWE(5 posts)→
More from Models
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11