Multimodal models put different demands on vision encoding, prefill and decoding. Separating vision encoding from the other stages can reduce resource contention and help models respond faster, but only for the right workloads. See how EPD disaggregation works, when it helps and what to consider before using it:

Multimodal reasoning has a latency problem. More video frames leads to more waiting. We built Damage Scout with Gemma 4 on Cerebras, running at over 2,300 toks/s, to show what fast multimodal inference unlocks. Damage Scout samples frames from a rental car walkaround, sends them to Gemma 4, gets back structured findings and box coordinates, then renders an annotated damage report in under 6 seconds. Same task. Same frames. A complete different experience powered by Cerebras ⚡️

A Multipolar World + $2T deficets in the USA + $2T AI Capex = Producer Prices - Business Level Inflation In Q1 / Q2, the BCOM vs. PPI divergence was striking; now the jaws are closing. - Bloomberg Terminal data.

A self-improving multimodal model can decide when to switch to fresh problems using only its training logs 🔄 A-RSI: Adaptive Task Refresh for Zero-Data Multimodal Self-Improvement 📌 Overview A-RSI is a method for refreshing the training task pool at the right time when a VLM improves itself without human labels. ⚠️ The problem ・Most work focuses on what to generate next, not when to replace the data ・Labels from majority voting can turn the model's own errors into training targets 🧩 Method ・The base model writes parameterized task generators that programmatically produce an image, a question, and a reference answer ・Training uses GRPO, with correctness ranked above all auxiliary rewards ・When prompts with reward variance drop or accuracy rises, the whole pool is replaced with unused tasks 📊 Results ・Qwen3-VL-8B's 11-benchmark average rose from 57.77 to 61.05 using 800 fresh tasks ・Refreshing every window reached 60.64 but used 3,200 tasks, four times as many ・MiMo-VL-7B also went from 59.08 to 60.71 🔭 Why it matters Timing the refresh without validation scores is practical. Note that each comparison is a single training run. #SelfImprovement# #MultimodalAI#

The specific terror of multipolarity, for those who grew up inside unipolarity, is not economic. It's existential. If America is not the center, what is it? If the dollar is not the reserve currency, what does that mean about the country that issued it? If other models of development exist and function, what does that say about the universal inevitability of the American one? These are not small questions for a population whose entire identity was built on the assumption that American primacy was the natural state of a correctly ordered world. The world reordering itself is experienced, from inside that assumption, as the world breaking. But the world is not breaking. The world is, for the first time in a long time, becoming more of what it actually is. Which is: not organized around any single center. Which is: billions of people with their own answers to questions America assumed it had solved.

This new-ish 35B multimodal model went from 3.38 → 20.51 tps on an iPhone. MiniCPM-V 4.7 35B-A3B was updated to decode faster on the iPhone by improving how it moves the model between storage and memory. Constraints 🧠 35B total / ~3B active 📱 iPhone 18 Pro Max — 12GB RAM 👁️ image + multi-image + video understanding Most of those 35B weights don't sit in RAM. The experts can stay in the phone's flash storage and get loaded when they're actually needed. Normally that's slow 🐌 SSD expert loading → 3.38 t/s So OpenBMB built Lookahead Expert Scheduling (LES) It predicts which experts the next layer will probably need and starts loading them while the current layer is still computing. Then they stacked the optimizations: 💾 baseline SSD loading → 3.38 tps 👀 expert lookahead → 6.57 ⚙️ optimized scheduling/loading → 11.34 🗜️ mixed-precision quantization → 17.16 🚀 + DSpark speculative decoding → 20.51 tps That's a 6.07× speedup over the same-phone unoptimized baseline. ⚠️ And these are still OpenBMB's own measurements. Link in ALT

Orbiting the multicolored lights of Southeast Asia. Here green lights used on fishing boats are in greater quantities than anywhere else on the planet, dense enough to appear as bright green streaks in long exposure photography.