Kimi K2.5
Kimi K2.5 by Moonshot AI: 11 offers. From €0.27 per 1M input tokens, with EU processing from €0.58.
Calculate costs for this modelCompare offersView price history
Profile
| Profile | |
|---|---|
| Developer | Moonshot AI |
| Type | Language models |
| Origin | CN |
| Input types | text |
| Context (tokens) | 256,000 |
| Max. output (tokens) | – |
| Reasoning | mit |
| Parameters (bn) | 1,000 |
| Open weights | ✓ yes |
| Licence | Modified MIT |
| VRAM at 4-bit (GB) | 554 (8 × RTX PRO 6000) |
| Released | – |
searchit Entity Extraction (–): 61.6 %
Offers with processing in the EU (1)
| Provider | Country | Variant | Input | Output | Source | As of |
|---|---|---|---|---|---|---|
| HostYourAI | NL | standard | €0.58 | €3.22 | hostyourai.com | 14 Oct 2026 |
€ per 1M tokens. Source: providers' pricing pages, read and checked by searchit. As of 14 Oct 2026.
Offers outside the EU (10)
| Provider | Country | Variant | Input | Output | Source | As of |
|---|---|---|---|---|---|---|
| AWS Bedrock | US | batch | €0.27 | €1.34 | pricing.us-east-1.amazonaws.com | 14 Oct 2026 |
| AWS Bedrock | US | flex | €0.27 | €1.34 | pricing.us-east-1.amazonaws.com | 14 Oct 2026 |
| SiliconFlow | SG | standard | €0.40 | €2.01 | www.siliconflow.com | 2 Oct 2026 |
| io.net | US | standard | €0.48 | €2.53 | api.intelligence.io.solutions | 14 Oct 2026 |
| AWS Bedrock | US | standard | €0.54 | €2.68 | pricing.us-east-1.amazonaws.com | 14 Oct 2026 |
| Cloudflare Workers AI | US | standard | €0.54 | €2.68 | developers.cloudflare.com | 10 Oct 2026 |
| Modular (Modular Cloud) | US | standard | €0.54 | €2.68 | www.modular.com | 14 Oct 2026 |
| Novita AI | US | standard | €0.54 | €2.68 | novita.ai | 13 Oct 2026 |
| Venice.ai | US | standard | €0.50 | €3.13 | docs.venice.ai | 14 Oct 2026 |
| AWS Bedrock | US | priority | €0.94 | €4.69 | pricing.us-east-1.amazonaws.com | 14 Oct 2026 |
€ per 1M tokens. Source: providers' pricing pages, read and checked by searchit. As of 14 Oct 2026.
AI Act
Whether the developer has signed the Code of Practice for general-purpose AI models: unknown. What this means is explained under “EU and AI Act”. EU processing and AI Act
Benchmarks
| Benchmarks | |
|---|---|
| GPQA Diamond (Artificial Analysis) | 87.9 |
| Humanity's Last Exam (Artificial Analysis) | 30.7 |
| ARC-AGI-2 (ARC Prize Foundation) | 11.81 |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 92.2 |
| SimpleQA Verified (Epoch AI) | 34.3 |
| SWE-bench Verified (Epoch AI) | 73.8 |
| Vectara HHEM-2.3 | 14.2 |
| Vectara HHEM-2.3 – antwortrate | 92.2 |
| Vectara HHEM-2.3 – halluzination | 14.2 |
| searchit Extraktions-Messlauf REPO-9 (PROMPT-CRAWLER-TEXT-START, 8 Dokumente × 3 Läufe) – treffer beziehungen | 47.4 |
| searchit Extraktions-Messlauf REPO-9 (PROMPT-CRAWLER-TEXT-START, 8 Dokumente × 3 Läufe) – treffer entitaeten | 75.8 |