Kimi K3 vs Fable Coding Comparison
ssh4net · x · 2026-07-19
Benchmarked on a real bug in Cline using the same harness: both Kimi K3 and Fable 5 can fix the issue, but with different strengths.
- Fable 5 is faster: total time 3.5 minutes, 18 tool calls; Kimi K3 took 12 minutes, 34 tool calls.
- Kimi K3 is cheaper: used 1.2M tokens, costing $0.92; Fable 5 used 730K tokens, costing $2.13.
- The author believes the traces show Kimi is more like a model trained with reinforcement learning, spending more tokens on thinking and verification.
- Conclusion: this is the first time they've seen an open-weight model genuinely compete head-to-head with SOTA in a real coding-agent scenario.
Related event: Kimi K3 Evaluations Show Polarized Results and Harness Sensitivity(5 posts)→
More from coding & agent
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11
- Codex tip: use Sol with Astra and Luna sub-agents to save usage — pvncher · 2026-09-11
- agents-best-practices: a provider-neutral Agent Skill for designing and auditing agentic harnesses — tom_doerr · 2026-09-11
- Cognition's SWE-2 uses a KKT duality argument in RL to shift the effort Pareto curve — YouJiacheng · 2026-09-11
- First-ever Three.js Conference lands in Paris, with a panel on AI-shortened design workflows — OdinLovis · 2026-09-11
- Agile co-author Ron Jeffries publishes 'Resist AI', urging developers to push back — mborch · 2026-09-11