I've been using both K2.6 and K2.7 for software engineering over the past few weeks, primarily through OpenCode.
K2.7 is a noticeable improvement for coding. It follows complex instructions better, produces cleaner code, and feels more reliable during agentic development workflows.
One area where it really impressed me was GitHub pull request code reviews. I tested it on a core TypeScript project, and the review quality was genuinely strong. It caught meaningful issues, provided useful suggestions, and understood the broader context of the changes instead of focusing only on superficial style comments.
The OpenCode integration is also excellent. The overall experience feels smooth and natural, making it easy to use K2 as part of a real development workflow.
When starting an entirely new project from scratch, I still think Claude Opus 4.8 has an advantage. However, for an open-source model, K2.7 exceeded my expectations by a wide margin. It's one of the most capable open models I've used, and I'll definitely keep using it.
Flowtica Scribe
Hi everyone!
Isn’t this pretty astonishing?
And the team is still being humble about it:
K3 is @Kimi’s 2.8T-parameter MoE model with 1M context and native multimodality.
It is built for long-horizon coding and knowledge work, and the results already put it directly into the frontier-model conversation. K3 can work across large repositories, use vision while iterating on games and frontend work, and keep going through much longer agent runs with limited supervision.
You can use it today on Kimi.com, Kimi Work, Kimi Code, or through the API.
Full weights arrive on July 27.
And the landscape may look very different after that..!
@zaczuo This is wild mainly because 1M context + native multimodality + strong coding changes what open model expectations are. A year ago that combo would have sounded like three separate roadmap items.
The full weights on July 27 are the part I’m watching. If teams can actually run or fine tune something this capable without handing everything to a closed API, the frontier model conversation gets a lot more interesting very quickly.
Kilo Code
@rauchg put it simply:
This is a breakthrough moment for open models. Three months ago, @Claude by Anthropic models were ahead for work requiring correctness and accuracy. Not anymore. LFG!
Kimi
We’re excited to introduce Kimi K3, our new open frontier model built for long-context, multimodal, agentic work!
Kimi K3 has 2.8T parameters, a 1M-token context window, and native multimodal capabilities. Under the hood, it introduces Kimi Delta Attention for much faster decoding in million-token contexts, plus Attention Residuals for more efficient training.
We built K3 especially for long-horizon coding, research, financial analysis, and self-evolving agent workflows — the kinds of tasks where models need to stay coherent, use tools, inspect outputs, recover from failed attempts, and keep going.
Kimi K3 is live now on Kimi.com, Kimi Work, Kimi Code, and the Kimi API.
Open weights are coming July 27, 2026.
Would love to hear what you build with it.
@crystal_j so incredibly excited for this! Congratulations on the launch. We can't wait to use it with PromptQL!
honestly the multi-file thing blew me away, dropped in like 20 mixed pdfs and slides and it actually pulled everything together coherently
K3 has a 1M-token context—what's the largest real-world task you've successfully completed end-to-end that simply wasn't practical with previous models?
Dropped a 40-page PDF into Kimi and asked it to compare it against a competitor's doc, and it pulled the right figures without me re-uploading anything. The slides maker is rough around the edges but shockingly fast for a first pass.
I'd love to see Kimi integrate with more file types, especially video and audio files. The real-time web search feature sounds incredibly useful, but I'm curious - how does it handle search queries with multiple keywords or complex phrases?