THE AI PULSEEN

The Pulse — August 2, 2026

The signals that entered our radar, organized with sources and context to understand what changed.

ModelsAgentsHardware
LISTEN TO THIS EDITION
The Pulse — August 2, 2026
Now in the PulseThis edition’s editorial thesis
RegusciLabs Pulse
Get the next Pulse directly.
  1. 01Google DeepMind — announcement

    Gemini Robotics 2: whole-body control, embodied reasoning, and on-device adaptation

    WHY IT ENTERED THE RADAR

    Google is splitting the robot stack into a VLA controller, an embodied-reasoning planner, and an on-device model. It claims new robot bodies can be adapted in a few hours with fewer than 200 examples, while ER 2 handles multi-minute, multi-step tasks and multi-robot coordination.

    SUGGESTED EDITORIAL ANGLE

    “AI robots just got a brain, a body, and an offline mode — why that three-layer stack is the important part.” Show planner → motion model → hardware as the visual.

    Open original source ↗
  2. 02Anthropic — Opus 5 announcement · Claude Code changelog (raw)

    Claude Opus 5 is now the default Opus model in Claude Code — with 1M context and a new fast mode

    WHY IT ENTERED THE RADAR

    Anthropic positions Opus 5 as near-Fable capability at half the price; the Claude Code release makes it the default Opus model, exposes a 1M-token context window, and lists $10/$50 per million input/output tokens for fast mode. The story is less “new benchmark winner” and more “long-running agent work is becoming a default developer workflow.”

    SUGGESTED EDITORIAL ANGLE

    “The real Opus 5 upgrade isn’t IQ. It’s that coding agents can hold an entire project in context — but should they?”

    Open original source ↗
  3. 03Google DeepMind — model page

    Gemini 3.6 Flash: efficiency becomes a competitive feature

    WHY IT ENTERED THE RADAR

    Google says 3.6 Flash uses 17% fewer output tokens than 3.5 Flash, while listing a 1M-token input window, 64k output, tool use and computer use. Its published table frames the model as a practical agent worker, not a cheap “lite” model.

    SUGGESTED EDITORIAL ANGLE

    “The next AI race is not only smarter models; it’s models that finish the same job with fewer tokens.” Explain why output-token efficiency directly affects latency and agent cost.

    Open original source ↗
  4. 04ByteDance Seed — Seedance 2.5 announcement

    Seedance 2.5 moves AI video from clips toward editable scenes

    WHY IT ENTERED THE RADAR

    Seedance 2.5 generates up to 30 seconds of synchronized audio-video per pass, supports multi-round extensions, and accepts up to 30 images, 10 video clips, and 10 audio clips as references. Timestamp-level editing and stronger reference control are the creator-relevant claims—not just prettier generations.

    SUGGESTED EDITORIAL ANGLE

    “AI video is becoming an editing system, not a prompt slot machine.” Demo the production workflow: references → one-take scene → targeted edit → extension.

    Open original source ↗
  5. 05Wafer — technical write-up

    Kimi K3 on AMD MI355X: memory capacity can beat the CUDA default for giant MoE inference

    WHY IT ENTERED THE RADAR

    Wafer reports Kimi K3 (2.8T parameters) reaching ~952 aggregate tok/s per MI355X node and 118 tok/s single stream. Its claimed performance-per-dollar advantage over B300 is driven by 288GB HBM per GPU and lower rental cost; it also shows how a small ROCm fix plus speculative decoding and an attention-kernel shape workaround materially changed results.

    SUGGESTED EDITORIAL ANGLE

    “The GPU winner for giant open models may not be NVIDIA.” Lead with the caveat: these are one vendor’s measurements, then explain why memory and prefill matter as much as headline decode speed.

    Open original source ↗
  6. 06FareedKhan-dev — kimi-k3-in-c

    A 2.78T Kimi K3 model running on one CPU with 8.24GB RAM — technically real, practically slow

    WHY IT ENTERED THE RADAR

    This C99 project streams routed experts directly from a 1.56TB checkpoint, keeps only a working set in RAM, and reports byte-identical output across memory budgets. The 8GB preset takes roughly 33 seconds per token: it is not a consumer serving recipe, but an unusually clear demonstration of how MoE architecture changes the memory problem.

    SUGGESTED EDITORIAL ANGLE

    “A 2.8-trillion-parameter AI on 8GB of RAM? Yes—at one word every half minute.” Explain the difference between fitting, running, and being usable.

    Open original source ↗
  7. 07NeelM0906 — Mference repository

    Mference brings SSD-streamed MoE inference to Apple Silicon

    WHY IT ENTERED THE RADAR

    The Swift/Metal runtime keeps the shared core and cache resident while streaming selected experts from SSD. It reports Gemma 4 26B at ~2GB memory and Qwen 3.6 35B at ~1.45GB, plus an experimental DeepSeek-V4-Flash path. The key practical signal: local inference tools are increasingly architecture-specific rather than generic wrappers.

    SUGGESTED EDITORIAL ANGLE

    “Your Mac’s RAM is no longer the full limit for local AI—but your SSD speed becomes part of the model.” Make the trade-off tangible: RAM, disk footprint, tokens/sec.

    Open original source ↗
  8. 08AI Builder Club — Open Agent Teams skill

    Open Agent Teams: a tiny protocol for reliable multi-agent coding workflows

    WHY IT ENTERED THE RADAR

    The upstream skill proposes detached tmux executors, a result file, and a file-based done sentinel instead of fragile terminal signals. This is the unglamorous operational layer that turns “many agents” from a demo into something observable and recoverable.

    SUGGESTED EDITORIAL ANGLE

    “Multi-agent workflows don’t fail because agents are dumb; they fail because orchestration is sloppy.” Show the simplest reliable pattern: coordinator, isolated worker, artifact, completion signal.

    Open original source ↗
TAKE THIS PULSE TO YOUR AI

Continue the analysis where you already work.

Copy this prompt into ChatGPT, Claude, Gemini, or whichever AI you use. It includes the signals, sources, and a guide for turning them into decisions.

No account is connected and no data is shared automatically.
PROMPT.md