AGENTKERNEL
- Home
- Leaderboard
- Full stack (agent)
Full stack (agent): the best AI models in October 2026
As of Oct 09, 2026, Claude Fable 5.1 leads the “Full stack (agent)” ranking, ahead of Kimi K3 and Muse Spark 1.3.
Best model with open weights: Kimi K3 (1,313).
Open in the interactive comparisonFull stack (agent) ranking
Elo rating from head-to-head user votes on generated results. Not comparable with index points.
| Rank | Model | Licence | Elo | Input / output per 1M |
|---|---|---|---|---|
| 1 | Claude Fable 5.1Anthropic | Proprietary | 1,325 | $10.00 / $50.00 |
| 2 | Kimi K3Moonshot AI | Open | 1,313 | $0.82 / $13.50 |
| 3 | Muse Spark 1.3Meta | Proprietary | 1,311 | $1.25 / $4.25 |
| 4 | Claude Opus 5Anthropic | Proprietary | 1,305 | $5.00 / $25.00 |
| 5 | Claude Fable 5Anthropic | Proprietary | 1,278 | $10.00 / $50.00 |
| 6 | Grok 4.6xAI | Proprietary | 1,263 | $2.00 / $6.00 |
| 7 | Claude Opus 4.8Anthropic | Proprietary | 1,248 | $5.00 / $25.00 |
| 8 | Claude Sonnet 5Anthropic | Proprietary | 1,248 | $2.00 / $10.00 |
| 9 | Grok 4.5xAI | Proprietary | 1,247 | $2.00 / $6.00 |
| 10 | Gemini 3.8 FlashGoogle | Proprietary | 1,242 | $0.75 / $3.75 |
| 11 | GLM 5.2Z.ai | Open | 1,239 | $0.06 / $6.00 |
| 12 | Claude Opus 4.6Anthropic | Proprietary | 1,218 | $5.00 / $25.00 |
| 13 | Claude Sonnet 4.6Anthropic | Proprietary | 1,211 | $3.00 / $15.00 |
| 14 | Muse Spark 1.2Meta | Proprietary | 1,208 | $1.25 / $4.25 |
| 15 | Gemini 3.7 FlashGoogle | Proprietary | 1,207 | $0.75 / $3.75 |