Chinese AI model reaches fourth on AI index

A new Chinese AI model has climbed to fourth place on the AI index, just behind the top proprietary models. This model, developed by Moonshot AI, shows significant improvements in agentic tasks, coding, and 3D modeling. The parameter count remains at 2.8 trillion, but API prices have dropped by 33 percent. Moonshot plans to release the model weights by November 2026.

The model scores 55 points on the Artificial Analysis Intelligence Index, an 11-point jump over its predecessor. Only three models rank higher, according to Artificial Analysis. If Moonshot releases the model weights as planned, it would become the highest-scoring open-weights model available. This would shrink the gap between open and proprietary models from 12 to just 3 points.

The model's strongest improvements show up in agentic tasks. On the GDPval-AA benchmark, it reaches 1,846 Elo, 312 points above its predecessor. On Terminal-Bench 4.0, Moonshot reports 65.7 percent, ahead of GPT-6 Astra. The model's predecessor scored only 13 percent on the same test.

Hallucination rates dropped sharply as well. On Moonshot's internal Kimi Hallu Bench Hard, claim-level hallucinations fell from 23.4 percent with the predecessor to 8.3 percent. The model also shows strength in financial tasks and general office tasks. It tops all tested models on OfficeQA Pro v2.

The model works directly with game engines like Unity, Unreal Engine, and Godot, handling both quick fixes and full project builds. On the BenchCAD benchmark, it scores 92.7 percent with tools, close to Claude Opus 5.5. The model can write and run scripts for Blender and CAD software without MCP connectors. The company also launched a feature called '3D Make' on Kimi.com that lets users preview, edit, and export 3D models.

The model keeps the same dimensions as its predecessor, with 2.8 trillion total parameters and 104 billion active ones. The context window spans one million tokens, and the model handles both text and images. Token prices have dropped by 33 percent. Input costs $2 per million tokens, output costs $10, and cache hits cost $0.20. Each Intelligence Index task costs $1.98.

Moonshot devotes an entire section of its technical blog post to cybersecurity. The company says it observed that vulnerability research capabilities kept growing as post-training scaled up. For the open-weights release, Moonshot stripped cyber-related data from both training phases. On its internal xdev-bench-hard, the uncleaned version solves 90.5 percent of tasks. After data cleaning, the model solves 81.0 percent. DeepSeek-V4.1-Flash reaches 85.7 percent on the same test, while the model's predecessor scored 33.3 percent.

Moonshot says no open-weights model should push the absolute frontier of offensive cyber capabilities. At the same time, the company wants to gradually raise what open models can do, since they also serve defensive purposes like auditing legacy systems. Anthropic previously showed that even GLM-5.3, which scores 45 on the AA Index, can write exploits at near-Claude Mythos levels.