Add corrections, implementation notes, pricing changes, or usage caveats for other readers.
Last updated
Jun 13, 2026
Input modalities
Output modalities
Capabilities
1,000,000 tokens
Recent tweets and retweets from Friendli
Heading to @icmlconf in Seoul? Come spend an evening with the teams building at the frontier.
FriendliAI is hosting an invite-only night on July 6 β the first day of ICML 2026.
π July 6, 2026 | PM β PM
π FriendliAI Seoul Office, Gangnam-gu, Seoul
π RSVP ββ¦
Deploy on FriendliAI to run Kimi K2.7 Code at agentic scale β where latency spikes and dropped requests aren't an option.
Run it on FriendliAI π friendli.ai/models/moonshotaβ¦
Link
moonshotai/Kimi-K2.7-Code API & Inference Endpoint | FriendliAI
Runβ¦
Kimi K2.7 Code is now live on FriendliAI.
@Kimi_Moonshot's coding-focused agentic model β built for long-horizon software engineering.
π +21.8% on Kimi Code Bench v2 vs K2.6
π€ ~10% stronger on agentic benchmarks
π§ 30% fewer thinking tokens
Article
Integrating FriendliAI with OpenClaw
Why OpenClaw + FriendliAI
Modern AI applications are no longer built around a single model call. Instead, they rely on agent systems that coordinate models, tools, and external services to complete
GLM-5.2 on FriendliAI: #1 throughput AND lowest tool call error rate on @OpenRouter.
β 48 tok/s avg β fastest of any provider
β 0.11% tool call error rate β the lowest on the board
Speed + reliability for long-horizon agents.
GLM-5.2 is live Day-0 on FriendliAI Model APIs:β¦
Discuss this model
Add corrections, implementation notes, pricing changes, or usage caveats for other readers.