Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Kimi-K3 Technical Report [pdf]

github.com

Post reply on HN
Print view View original

1–10 of 199 posts

Re: Kimi-K3 Technical Report [pdf]

#2

Post by m00dy »

I would want to see three things before drawing strong conclusions:

End-to-end tokens/sec and cost on realistic coding agent trajectories, including tool outputs and retries, not isolated decode benchmarks.

Cache hit rates and prefill cost for branching, multi-turn sessions.

Router-load distributions after post-training, where expert collapse or specialization problems often show up.

Re: Kimi-K3 Technical Report [pdf]

#4

Post by whimsicalism »

storus wrote: post #3

What would be the current best method to fine-tune it for my own specific agentic tasks? LoRA + DPO? GRPO? Something else?

LoRA + SFT, but it'll be big - better to wait for a finetuning API from one of the providers, I wouldn't jump straight to RL or off-policy pseudo-RL like DPO.

Re: Kimi-K3 Technical Report [pdf]

#6

Post by jvanderbot »

m_ke wrote: post #5

Also open sourced a bunch of infra to go with it. Anyone who claims open source and open weights models are "decel" needs to get their head checked https://github.com/MoonshotAI/MoonEP https://github.com/kvcache-ai/AgentEnv https://github.com/MoonshotAI/FlashKDA

This comment would be much better without the second line

Re: Kimi-K3 Technical Report [pdf]

#8

Post by fahrradflucht »

License: https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE

> If the Licensee or any of its affiliates operates a Model as a Service business, and the aggregate revenue of the Licensee and its affiliates exceeds 20 million US dollars (or the equivalent in other currencies) in total over any consecutive 12 months, the Licensee must enter into a separate agreement with Moonshot AI before using the Software or its derivative works for any commercial purpose.

+ the existing 100 million monthly active users, or more than 20 million US dollars for commercial products have to name Kimi clause

Re: Kimi-K3 Technical Report [pdf]

#9

Post by ike_a »

jvanderbot wrote: post #6
m_ke wrote: post #5

Also open sourced a bunch of infra to go with it. Anyone who claims open source and open weights models are "decel" needs to get their head checked https://github.com/MoonshotAI/MoonEP https://github.com/kvcache-ai/AgentEnv https://github.com/MoonshotAI/FlashKDA

This comment would be much better without the second line

I'm not sure I understand the case for open-source models being decelerationist, is this it?

Decel:

- Potentially reduces investor appetite for funding big labs.

- More risk of powerful AI getting in bad hands -> more regulation.

Accel:

- More competition so big labs can't rest on laurels.

- More research in open, so all labs can accrete advancements faster.

I feel like open-source = acceleration has a much more clear argument. (and how bad would deceleration be in any case?)

Re: Kimi-K3 Technical Report [pdf]

#10

Post by Der_Einzige »

jvanderbot wrote: post #6
m_ke wrote: post #5

Also open sourced a bunch of infra to go with it. Anyone who claims open source and open weights models are "decel" needs to get their head checked https://github.com/MoonshotAI/MoonEP https://github.com/kvcache-ai/AgentEnv https://github.com/MoonshotAI/FlashKDA

This comment would be much better without the second line

[flagged]
Post reply on HN