Kimi K2.6: what developers actually say
We read every Hacker News comment from Nov 2025 – Nov 2026 that gave a real opinion on Kimi K2.6. 31 of the 38 opinions come from people who say they use it themselves.
Kimi K2.6 scores 70% approval across 38 opinions from 31 authors, and 31 of those opinions are firsthand. Code quality is the strongest signal at 11 positive to 3 negative, and cost is 8 positive to 1 negative. Approval fell from 90 in April to 58 in May, then moved to 64 in June and 75 in July, though July has only 4 opinions. Speed and reliability draw more criticism than praise, with reliability at 1 positive to 2 negative.
Pick it if
- Need solid code on tightly scoped tasks
- Already have a detailed plan from a Claude model
- Do security or binary analysis work
Look elsewhere if
- Write Zig
- Run long agentic workflows
- Need fast local inference
Strengths and weaknesses
Each row counts opinions that mentioned the topic. Bars to the right of the line are praise, bars to the left are complaints.
How opinion has moved
What developers say
Praise
Complaints
Summaries are paraphrased. Each links to the original comment.
FAQ
Is Kimi K2.6 worth it?
It holds 70% approval across 38 opinions. Code quality runs 11 positive to 3 negative and cost runs 8 positive to 1 negative. Reliability is the weak spot at 1 positive to 2 negative, so the value depends on how much you need long runs to hold up.
Is Kimi K2.6 better than Claude Opus?
Several developers put its code output close to Opus, and one says the gap nearly closes when Opus writes the plan. One person preferred it over Opus 4.8. Others say it needs more correction than Opus on the same work.
Is Kimi K2.6 good for agentic coding?
Opinions are split. Some developers say it loses track during long agentic workflows, and one got no good results on local agentic tasks. Others say it follows Claude-made plans reliably and keeps spend low.