AI Model Comparison
Side-by-side comparison of pricing, context window, benchmarks, capabilities, and license. Every figure comes from the developer's own pages, last checked on ; each model's page lists its sources. A figure we could not source shows as a dash.
Model A — Swiss AI Initiative
Apertus 1.5 70B is the 70B-parameter member of Apertus 1.5, the fully open multilingual model family from the Swiss AI Initiative (ETH Zurich, EPFL and CSCS), released on July 24, 2026. It was continued-pretrained from Apertus 1.0, takes image, audio and text input, generates text, and has an optional thinking mode.
Model B — Moonshot AI
Kimi K3 is Moonshot AI's 2.8-trillion-parameter open-weight Mixture-of-Experts model with native vision and a 1-million-token context window, built on Kimi Delta Attention and Attention Residuals. On the API it always reasons, with configurable reasoning effort.
| Property | Apertus 1.5 70B | Kimi K3 |
|---|---|---|
| Released | Jul 24, 2026 | Jul 16, 2026 |
| Developer | Swiss AI Initiative | Moonshot AI |
| Type | multimodal | multimodal |
| License | open-source | open-weight |
| Context window | 262,144 | 1,048,576 |
| Parameters | 70B | 2.8T total, 104B activated (MoE; 16 of 896 experts per token) |
| Developer status | — | — |
| Input $/M tokens | — | $3.00 |
| Output $/M tokens | — | $15.00 |
Bold green values indicate the winner on that dimension where comparison is meaningful (lower price wins; bigger context wins; more open license wins; newer date wins). Prices are the ones each developer lists today; a model it no longer prices has none.
Programmatic access available at /api/ai/compare?a=apertus-1-5-70b&b=kimi-k3. CORS is open. Citation guidance: /about/editorial-process.