SiliconFlow gives Kimi K3 a 20× TPM boost and a 10% price cut
SiliconFlow has increased Kimi K3's default TPM from 100K to 2M (20×) and cut prices by 10% for all users, lowering cache input, input, and output rates. The ne…
SiliconFlow has increased Kimi K3's default TPM from 100K to 2M (20×) and cut prices by 10% for all users, lowering cache input, input, and output rates. The ne…
SiliconFlow announces a major upgrade for Kimi K3: 10% cheaper pricing and a 20× increase in default tokens per minute (TPM), from 100K to 2M, with new rates fo…
MiniMax is inviting the AI community to vote for the best acceleration version of its H3 model, with links to the poll and related details.
Tencent Hunyuan's Hy4 Preview, a 770B-parameter sparse MoE open-weights model activating 49B parameters per token, is now available via official API. It scored…
Tencent Hunyuan's Hy4 Preview, a 770B-parameter sparse MoE open-source model with 49B active parameters per token, is now available via official API access, per…
Sam Altman highlighted a post arguing that frontier AI model depth has only grown modestly since GPT-4, and warned against confused reporting that could trigger…
MiniMax shares H3-World, a new approach that turns the MiniMax-H3 language model into a world model using only 8K samples and about 0.199% trainable parameters.
Starting Sept 3, 2026 at 17:00 SGT, DeepSeek V4 Flash and V4 Flash Vision Exp will leave the free tier and adopt discounted pricing, with 50% off during peak ho…
B.AI announces DeepSeek V4 Flash and V4 Flash Vision Exp will leave the free tier and switch to discounted billing on September 3, 2026. The mysterious Ox Alpha…
Command Code CEO Ahmad Awais announced that a new free AI model will launch in the platform tomorrow morning, following earlier hints about shipping Alibaba Qwe…