Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-05-25 · source-backed
Bloomberg reports DeepSeek will permanently maintain the V4-Pro discount that was set to expire end of May. Input pricing from $1.74 to $0.435/M tokens, output from $3.48 to $0.87. Enabled by migration to Huawei Ascend 950 accelerators and the 1.6T-parameter MoE architecture activating only 49B during inference.
Each link below shares sources, entities, or timing with this story.
DeepSelect is the TopK kernel behind the indexer in DeepSeek Sparse Attention, and DeepSeek says it runs 2-20x faster than torch.topk. DeepJIT is a header-only C++20 runtime that compiles kernels at runtime, with one interface for both NVIDIA CUDA and Huawei Ascend. deepseek-r...
For about a year, "run your agent locally" meant accepting a model that couldn't reliably call a tool twice in a row. That excuse is gone. Meta Superintelligence Labs published Muse Glimmer today: a 29.6B dense causal transformer, 52 layers, 6,656 hidden dim, with a ~1.8B ViT-...
DeepSeek released V4 on April 24 and the numbers demand attention. V4-Pro is 1.6 trillion parameters total with 49 billion active, MIT-licensed, native 1M-token context. It scores 80.6% on SWE-bench Verified, putting it within 0.2 points of Claude Opus 4.6. On Terminal-Bench 2...
DeepSeek released V4 Preview on April 24 with two open-weight variants: V4-Pro (1.6T total parameters, 49B activated via MoE) and V4-Flash (284B parameters, 13B activated). Both support 1M-token context windows. Both are Apache 2.0 licensed. Both are live right now on Hugging...
GLM-5.1 from Zhipu AI scored 58.4% on SWE-bench Pro. GPT-5.4 scored 57.7%. Claude Opus 4.6 scored 57.3%. That's the first time an open-weight model has ever topped a major coding benchmark against the best proprietary models. The specs matter. GLM-5.1 is a 754B-parameter mixtu...
The assumption that proprietary models own the coding benchmark crown just broke. Moonshot AI's Kimi K2.6 leads on 5 of 8 major agentic coding benchmarks while being the only open-weight model in the top tier. SWE-Bench Pro: 58.6% vs GPT-5.4's 57.7% and Claude Opus 4.6's 53.4%...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.