Venice Adds Private Access to Moonshot's Kimi K3 With US Hosting and Zero Data Retention
Venice said it has added private access to Moonshot's Kimi K3. Erik Voorhees said the service runs from a US data center with zero-data retention.
Voorhees said Kimi K3, an open model trained by Moonshot, can be run by providers other than Moonshot, so users accessing it through Venice are not using Moonshot's infrastructure. He also said Moonshot's own service applies a separate company filter to some queries and that Venice does not add that layer.
From the sources (25 posts)
@modalKimi K3 drops tomorrow. Day 0 support on Modal.
@teortaxestexmeanwhile time in China: 00:32 Monday, July 27, 2026 Strictly under 24 hours until an open frontier model
@kimmonismusQuick reminder on Kimi k3 open weights
@aaazzamRT @modal: Kimi K3 drops tomorrow. Day 0 support on Modal.
@togethercomputeKimi K3 lands on Together tomorrow. Available on Provisioned Throughput: reserved token-based capacity that just works: 1/ Guaranteed tok/min
@businessMoonshot AI is poised to make its Kimi K3 model available for public download, expanding its reach in the global open software community at a time of growing US concern about Chinese encroachment into the top echelons of AI development http
@cointelegraph🇨🇳 JUST IN: China's Moonshot AI is set to release its Kimi K3 model weights for free public download, widening its reach in the open-source AI race against US rivals.
@brianroemmeleLocal Kimi K-3 download day in just a few hours! I have 16 very large US companies ready to download and run it of Hopper class GPUs. All but two have staff on night watch to get it done. There is a hidden Anthropic Easter Egg on the site
@businessMoonshot AI is poised to make its Kimi K3 model available for public download, expanding its reach and influence in the global open software community at a time of growing US concern about Chinese encroachment into the top echelons of AI de
@nrehiew_Happy K3 weights and paper release day for all who celebrate
@jun_song8 hours left for Kimi-K3 weight release. Where could I get another 2TB of VRAM in 8 hours? All the assets I have left are just two kidneys
@tekniumK3 will be open sourced in 8 hours. Woot!
@kimi_moonshotReleasing the model weights and technical report of Kimi K3. Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window. New model architecture: 2.5x the intelligence per unit of com
@wesrothKimi K3’s full model weights will be released by July 27, 2026. Architecture details, training methods, and evaluation results will follow in the technical report.
@rgb_lightsCareful with AI model performance claims at launch. First reports are usually wrong. “Kimi K3 performs significantly below the most recent frontier cyber-capable models on preliminary cyber evaluationsrun by UK AISI / CAISI.” https://t.c
@teortaxestexEuropean* excellence. Kimi K3 is live on eu-north1 at Nebius. And yes of course fp4. Same cost, 3-4 times faster. *formerly Yandex
@kimi_moonshotKimi K3 (open weights, coming soon)
@lentils80🚨 Kimi K3 is now finally open source and weights are out on Hugging Face!
@scaling01Kimi K3 is now officially open-weights
@basetenKimi K3 is now live on our Model APIs, day 0.
@basetenWe’re thrilled to be day 0 launch partners with the @Kimi_Moonshot team. Try it now:
@andrewcurran_K3 weights and technical report have been released.
@crystalsssupRT @Kimi_Moonshot: Releasing the model weights and technical report of Kimi K3. Kimi K3 is our most capable model: a 2.8T MoE model with n…
@victormustarRT @Kimi_Moonshot: Releasing the model weights and technical report of Kimi K3. Kimi K3 is our most capable model: a 2.8T MoE model with n…
@kimmonismusMoonshot delivered: model weights and technical report of K3 are now live!