[Tinygrad Claims GLM 5.2 Can Achieve 120 tok/s on Dual-Connected Blackwell Configuration, Priced at $150,000] BlockBeats News, June 21: GPU vendor Tinygrad announced that, according to reliable sources, the GLM 5.2 model can achieve an inference speed of 120 tokens per second on two networked Blackwell architecture tinyboxes. This configuration is priced at $150,000 and offers the option of either two standard tinyboxes or a single tinybox Pro, both capable of delivering the aforementioned performance. Tinygrad is positioning this as a selling point, emphasizing a "one-time purchase, no cloud fees ever" private deployment model, directly competing with pay-as-you-go cloud inference services. Currently, this information has not been confirmed by GLM officials, and Tinygrad has not disclosed further technical details. --------------------------------- Click the link below to access the original article and join the Beating · Feishu AI News Channel for 24/7 global AI news and updates. [Original Link]