Grok 4.5 Ranks Second in Coding Benchmark
XFreeze · x · 2026-07-19
Grok 4.5 ranked 2nd on AlphaSignal's private SignalDesk V1 coding-agent benchmark, successfully resolving 69 out of 70 attempts for a 99% resolve rate.
The post also highlights its efficiency: the cost per successful fix is about $0.074, with a total cost of $5.12, an average completion time of 46 秒, and a total consumption of 510 万 tokens. The author concludes that Grok 4.5 is not only highly capable but potentially one of the most cost-effective coding agents available right now.
More from Models
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- ChatGPT tells user their question is too hard and to 'accept dumber answers' — phido3000 · 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