FrontierThe story, in brief
Qwen3.8 Max catches Claude Opus 4.8 but Kimi K3 still scores higher for 25 percent less
Qwen3.8 Max just closed the gap with Claude — but Kimi K3 is still winning at 25% lower cost.
The KeyNews take
Why it matters
The frontier model race is tightening: Alibaba's latest closes on Anthropic's Opus 4.8, while Moonshot's Kimi K3 remains the efficiency leader. This reshapes the competitive calculus for practitioners choosing which API to standardize on.
The key facts
5 to knowQwen3.8 Max scores 56 on Artificial Analysis Intelligence Index (10-point jump from Qwen3.7 Max's 46)
Qwen3.8 Max now competitive with Claude Opus 4.8
Kimi K3 still scores higher than both, with 25% lower cost
Benchmark: Artificial Analysis Intelligence Index (capability measurement)
Vendors in race: Alibaba (Qwen), Anthropic (Claude), Moonshot (Kimi)
Go to the source
The Decoderthe-decoder.com
Publisher excerpt: Alibaba's Qwen3.8 Max scores 56 on the Artificial Analysis Intelligence Index, a 10-point jump over Qwen3.7 Max (46).