Muse Spark 1.1 Scores 69 in Coding Benchmark
ArtificialAnlys · x · 2026-07-12
According to Artificial Analysis, Muse Spark 1.1 (xhigh) scored 69 on their Coding Agent Index, performing near the frontier under the Opencode harness with strong cost efficiency. For context, it scored slightly lower than GPT-5.5 (medium) at 71, but higher than Claude Opus 4.8 (medium) at 67. The post also notes that its cost per task is roughly $1.4, making it one of the cheaper frontier coding agents, though this comes with certain capability trade-offs.
Related event: Meta Muse Spark 1.1 Shines Across Multiple Benchmarks(13 posts)→
More from coding & agent
- Goal-driven AI needs verifiable success signals, or it invents its own — daniel_mac8 · 2026-09-11
- Frontier models need ways to verify success — or they'll invent their own — daniel_mac8 · 2026-09-11
- Sakana AI launches Fugu Max: dynamic multi-agent routing across its largest open-model pool — graceisford · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- Anthropic researcher: 99% of engineers now run swarms of 300+ self-improving agents — AlishaOutridge · 2026-09-11
- Gergely Orosz: Shipping 10x PRs With AI Agents, Sites Fill With Small Regressions — ducha_aiki · 2026-09-11