Moonshot AI Narrows the Gap With U.S. Rivals in New Kimi Release

Moonshot AI has officially launched Kimi K3.1 Pro, a significant upgrade to its reasoning capabilities that places the Chinese firm within striking distance of top-tier American competitors. As of late October 2026, the new model has secured the fourth spot on the Artificial Analysis Intelligence Index, marking a notable surge in performance that challenges the dominance of proprietary systems.

The numerical leap is substantial. Kimi K3.1 Pro scores 55 points on the Intelligence Index, an eleven-point increase over its predecessor, Kimi K3. This score sits just three points behind Claude Opus 5.5, which leads the pack with 58 points. Claude Sonnet 5.5 and Muse Wonder Full also rank higher, both scoring 56. If Moonshot follows through on its plan to release the model’s weights by November 4, 2026, K3.1 Pro will become the highest-scoring open-weights model available, narrowing the performance gap between open and proprietary architectures from twelve points to just three.

The company is also distributing a variant called K3.1 Pro Highspeed, which Moonshot claims runs two to three times faster than the standard version. Both models are currently accessible through Kimi.ai, Kimi Work, Kimi Code, and the Kimi API, offering developers and businesses immediate access to the updated capabilities without waiting for a formal open-weights launch.

The most dramatic improvements are visible in agentic tasks, where the model executes complex, multi-step operations. On the GDPval-AA benchmark, K3.1 Pro reaches an Elo rating of 1,846, a 312-point rise from K3. This places it slightly behind Claude Opus 5.5 (1,866) and Muse Wonder Full (1,857). In coding environments, the model achieves 65.7 percent on Terminal-Bench 4.0, surpassing GPT-6 Astra’s 58.2 percent and trailing only Claude Opus 5.5 at 66.4 percent. By comparison, the previous K3 model managed only 13 percent on that same test.

Reliability has also improved, with hallucination rates dropping sharply. According to Moonshot’s internal Kimi Hallu Bench Hard, claim-level hallucinations fell from 23.4 percent in K3 to 8.3 percent in the new version. While GPT-6 Astra maintains a lower rate of 5.5 percent, Moonshot notes that its benchmark is a stress test rather than a general measure of error. In finance, the model gained 7.2 points on Financial Modeling and 11.6 points on Financial Research, while topping all tested models on OfficeQA Pro v2 for general office tasks.

Beyond traditional coding, K3.1 Pro demonstrates strong capabilities in 3D modeling, working directly with game engines like Unity, Unreal Engine, and Godot. On the BenchCAD benchmark, it scores 92.7 percent with tools, close to the leaders, and 72.8 percent without tools, nearly matching Claude Opus 5.5. The company has also launched a "3D Make" feature on Kimi.com, allowing users to preview, edit, and export 3D models, with the model excelling at parametric modeling of mechanical and architectural parts.

Despite the performance gains, the model retains the same architecture as K3, with 2.8 trillion total parameters and 104 billion active parameters. It supports a one-million-token context window and handles both text and images. However, the cost of using the model has decreased significantly. Input prices drop to $2 per million tokens from $3, output prices fall to $10 from $15, and cache hits cost $0.20 down from $0.30. This represents a 33 percent reduction in API pricing, keeping the cost per Intelligence Index task at $1.98, nearly identical to the previous version.

Moonshot has also addressed cybersecurity concerns by removing cyber-related data from the open-weights version to balance offensive and defensive capabilities. On its internal xdev-bench-hard, the uncleaned model solved 90.5 percent of vulnerability tasks, while the cleaned K3.1 Pro solved 81.0 percent. This approach aims to ensure open models serve defensive purposes without pushing the absolute frontier of offensive cyber capabilities, a balance the company believes is essential as open models become more powerful.

The launch signals a tightening race in the AI sector, with Moonshot positioning itself as a serious competitor to established giants. By combining significant performance gains with lower costs and a commitment to open weights, the company is aiming to reshape the landscape for developers and enterprises relying on advanced AI models.