#Deployment/Engineering

AIAntigravity has released an update that shows hourly and weekly model quota in percentage increments, with the dashboard coming soon to the IDE. The company also reset Gemini quota for all users.

AIOpenRouter introduced Fusion, a tool that sends a prompt to a panel of models and has a judge model fuse their results into one answer. On 100 DRACO deep research tasks, a Fable 5 and GPT-5.5 panel scored 69.0%, above Fable 5 alone at 65.3%, and a budget panel of Gemini 3 Flash, Kimi K2.6, and DeepSeek V4 Pro reached 64.7% at about half the cost of Fable 5.

Why it matters: The source gives benchmark scores, panel compositions, and contamination controls, letting readers judge how much of the gain comes from model diversity versus self-synthesis.

AIFactory is rolling out automated security review in Droid, running a STRIDE-based check on every non-draft PR alongside standard code review. Findings include severity, a CWE reference, an explanation, and a suggested fix, posted as inline diff comments. The feature is available today on all plans, and a deeper multi-agent /security-review deep audit is available for full-repository scans.

AIZed is building DeltaDB, a version control system that records every operation as a fine-grained delta, linking agent conversations to the code they produce. The company says a beta will arrive in a few weeks, and it invites users to join a waitlist. The system is designed so teammates can collaborate on work in progress without waiting for commits, pull requests, or pushes.

AIXiaomi's MiMo team released MiMo Code, an MIT-licensed terminal coding agent built on OpenCode for long-horizon programming tasks. The design centers on three areas: Max Mode parallel sampling that generates five candidates per turn, Goal-based completion verification, and a memory system that checkpoints session state and rebuilds context. The article reports offline benchmark results and a double-blind A/B test with 1,213 pairs in which MiMo Code's win rate exceeded 65% beyond 200 execution steps.

Why it matters: The article explains how MiMo Code handles long-horizon coding through computation, checkpointed memory, and cross-session evolution, useful for judging design tradeoffs in coding agents.

AIApple's MLX team released three videos at WWDC covering running agents locally, distributed inference and training, and MLX Swift. The post shares YouTube links for each talk, presented by Angelos Katharopoulos, Tatiana Likhomanenko, and David Koski.

AICognition introduced FrontierCode, a coding benchmark built with open-source maintainers that measures whether models produce code a maintainer would merge. On FrontierCode Diamond, the hardest 50 tasks, Claude Opus 4.8 scored 13.4%, GPT-5.5 scored 6.3%, and Gemini 3.1 Pro scored 4.7%. The authors report 81% fewer misclassification errors than SWE-Bench Pro, though this figure comes from their own analysis of agent trajectories.

Why it matters: The benchmark's blocker and rubric design shows how code quality can be measured beyond unit-test correctness, which matters for judging coding agents.

AIXiaomi MiMo and the TileRT team say a 1T model exceeds 1,000 tps on a single standard 8-GPU node using general-purpose GPUs. The speedup comes from FP4 quantization and DFlash, a block-masked parallel speculative decoding method that accepts more tokens per verification, with TileRT tailoring its compiler and kernels to these techniques. Open weights for the FP4 + DFlash checkpoint are available on Hugging Face.

AIXiaomi MiMo has released MiMo-V2.5-Pro-FP4-DFlash, the FP4 backbone behind MiMo-V2.5-Pro-UltraSpeed, with MXFP4 quantization applied only to the MoE experts and a BF16 DFlash drafter for block-diffusion speculative decoding. The backbone has 1.02T total and 42B active parameters, and the drafter proposes blocks of up to 8 tokens per forward pass. The release is supported in SGLang, with example launch commands provided.

AIXiaomi and TileRT released MiMo-V2.5-Pro-UltraSpeed, reporting decode speeds above 1000 tokens/s on a 1-trillion-parameter model using a single standard 8-GPU node. The API is priced at 3x MiMo-V2.5-Pro and is available by application only from June 9 to June 23, 2026. The speedup relies on FP4 quantization of MoE Experts, DFlash speculative decoding with an average coding acceptance length of 6.30, and TileRT compute kernels.

Why it matters: The post traces how FP4 quantization, DFlash speculative decoding, and TileRT kernels combine to reach 1000+ tokens/s on a single 8-GPU node, which is useful for teams weighing inference throughput.

AISebastian Raschka has published a curated list of LLM research papers he bookmarked from January through May 2026, not a complete survey of the field. The list is weighted toward reasoning models, reinforcement learning, and efficient inference, with added interest in agent harnesses, long context, and diffusion language models. He highlights Nvidia's Nemotron 3 Super, a 120B-A12B hybrid model alternating attention and Mamba-2 layers, as a must-read, and notes a 4B Nano variant and the 550B-A55B Nemotron 3 Ultra released two days earlier.

AICohere and Cohere Labs released North Mini Code 1.0, an open-weights 30B-A3B mixture-of-experts model for code generation and agentic terminal tasks, under Apache 2.0. The model has 256K context and 64K max output, and is trained for tool use. Its benchmark table lists Terminal-Bench v2 at 36.0, SWE-Bench Verified at 67.6, and LiveCodeBench v6 at 70.3, below Qwen3.6 on several tasks.

Why it matters: The card lists benchmark results against Qwen3.6, Gemma4, and other models, showing where North Mini Code trails on some coding and agentic tasks.

AIAntigravity has enabled the /teamwork-preview command for all paid plans, letting users run parallel implementation and verification agents on complex tasks. The post says the team built a working OS with it, but warns that it can consume a large number of tokens.

AIBaidu CoBuddy is now available for free on Novita AI, a code-focused model aimed at developers and AI agents. It offers a 131K context window, up to 65K output tokens, and native tool calling. The model is served through Novita's serverless API for high-throughput, low-latency inference.

AICognition built an automated agent that classifies Devin sessions as productive and estimates the human engineering hours each one would have taken. On 233 held-out sessions the estimator reached an rlog of 0.74, with individual errors often 2 to 3 times in either direction but roughly unbiased in aggregate. The system is calibrated to underestimate and is currently running with Devin customers.

Why it matters: The post shows how the measurement design, from hours-based metrics to conservative calibration, determines whether agent productivity estimates can be trusted in aggregate.

AIGeorgi Gerganov says Computex this year shows a strong signal for local AI, with major players like NVIDIA and Microsoft embracing and discussing local AI workloads. He adds that dedicated consumer hardware and models are on the way.

AIMiniMax published MiniMax-M3-MXFP8, an MXFP8 quantized variant of its native multimodal M3 model with 1M context, about 428B total parameters and about 23B activated parameters. M3 adds MiniMax Sparse Attention, which the source says yields 9× prefill and 15× decode speedups over M2 at 1M context. The model supports three thinking modes (enabled, adaptive, disabled) via the thinking parameter and can be served with SGLang, vLLM, or Transformers.

Why it matters: The release pairs sparse attention for 1M-token contexts with reported prefill and decode speedups over M2, useful for judging long-context serving costs.

AIXiaomi MiMo has published a blog detailing full-pipeline inference optimizations for the MiMo-V2.5 series. The post highlights how the team pushed hybrid sliding window attention (SWA) efficiency to its limit. The full write-up is available at

AIXiaomi describes an end-to-end inference optimization for the MiMo-V2.5 series, centered on Hybrid SWA, which it says cuts KVCache storage to roughly 1/7 of Full Attention. The post covers a dual KVCache pool design, SWA-aware prefix cache matching, the GCache distributed cache, and scheduling changes, and reports cache hit rates averaging 93% in server-side observations. It also covers prefill and decode optimizations, multimodal encoder improvements, and open-source contributions to SGLang.

Why it matters: The post explains how Hybrid SWA's theoretical KVCache savings were realized in production through dual pools, SWA-aware prefix caching, and tiered storage, giving concrete engineering patterns for long-context inference.

AICognition describes autonomous testing in Devin, where the agent writes a source-grounded test plan, operates the app through computer use, and returns labeled screenshots and an annotated video. Login steps are handled by a deterministic testing skill, and the company says test runs approved per day more than doubled in recent months. Known limits include timing errors with transient UI elements and models sometimes triggering states through JavaScript instead of clicking the interface.

Why it matters: The post explains how computer use, test plans, deterministic login scripts, and annotated recordings let Devin verify its own code changes end to end.

AICognition raised over $1B at a $26B valuation, led by Lux Capital, General Catalyst, and 8VC. Enterprise Devin usage grew more than 10x since the start of this year, and run-rate revenue reached $492M. The company also said it launched SWE-1.6, which reaches up to 950 tok/s and has become the most used model in Devin Desktop.

AIMiniMax has upgraded its Agent, renamed Mavis, and introduced Agent Teams that run multiple role-based Agents in parallel on desktop. The team uses Leader, Worker, and Verifier roles so complex tasks can be split, checked, and reported at key checkpoints, and it merges TokenPlan and Agent Plan into one subscription with credits shared between Agent and API. The post also discusses the added token, handoff, and retry costs of multi-Agent work, and says the Agent will be open-sourced alongside MiniMax M3.

Why it matters: The post explains why multi-Agent helps long tasks and where its verification, token, and aggregation costs come from, useful for judging when a team setup beats a single Agent.

AIMiniMax says its M2 series could not output the name Ma Jiaqi, a failure it traced to post-training data that rarely included the token. Its tests found the input embedding stayed stable while the lm_head weights for low-frequency tokens drifted during SFT. A synthetic full-vocabulary repetition dataset restored generation for affected tokens and reduced Japanese-to-Russian confusion from 47% to 1%.

Why it matters: The post traces a community-noticed token failure through tokenizer, embedding, and lm_head tests, showing how post-training data coverage can cause low-frequency token drift.

AIFunAudioLLM has released Fun-ASR-Nano-2512-hf, a Hugging Face Transformers-compatible version of its end-to-end speech recognition model, which supports Chinese, English, and Japanese. The Chinese coverage includes 7 dialect groups and 26 regional accents, and a separate Fun-ASR-MLT-Nano-2512 checkpoint handles 31-language recognition. Developers can run the model natively in Transformers 5.17.0 without custom model code or trust_remote_code=True.

AITri Dao says that after a mathematical rewrite, all transformer operations can be expressed as a series of GEMMs with epilogues. Given a few optimized primitives, LLMs and novice humans can write near speed-of-light kernels for transformer ops. The related CODA work fuses memory-bound surrounding ops into the matmul epilogue, and LLMs can also write CODA kernels approaching speed-of-light.

AICursor is doubling usage for new users invited on the Teams plan for the next month. The offer applies only to users invited by a Teams plan member during that period.

AICognition's Devin AI software engineer can now build, run, and test code natively in its own Windows virtual machine, including migrating .NET Framework apps to .NET Core. The Windows capability is in beta for Enterprise Cloud and Dedicated Deployment customers, with the same SOC 2 Type II and ISO 27001 controls as the Linux version. Citi and Mercedes-Benz are named as existing Devin users.

AIPaddleOCR 3.5 now supports Hugging Face Transformers as an inference backend, letting users run PP-OCRv5 and PaddleOCR-VL 1.5 models directly within the Transformers ecosystem. Users can select it with engine="transformers" while keeping the same PaddleOCR pipeline, which the post says eases integration for RAG and Document AI applications.

AICognition has released Auto-Triage in Devin Automations, which lets Devin respond to Slack messages, Linear events, GitHub activity, schedules, and webhooks. Devin can investigate with connected observability tools and the codebase, then post a summary, tag an owner, or open a PR. Devin runs in network-sandboxed environments with added protections against prompt injection and data exfiltration, and a limited-time offer gives $200 in credits for a first automation.

Why it matters: The post shows how an agent handles alerts and bug reports from existing team channels, a practical pattern for teams weighing automated incident response.

AISebastian Raschka reviews recent open-weight LLM architecture changes aimed at reducing long-context memory and compute costs. He covers KV sharing and per-layer embeddings in Gemma 4, per-layer query-head budgeting in Laguna XS.2, Compressed Convolutional Attention in ZAYA1-8B, and mHC with CSA/HCA compressed attention in DeepSeek V4. The article reports that DeepSeek V4-Pro uses 27% of single-token inference FLOPs and 10% of the KV cache size of DeepSeek V3.2 at a 1M-token context.

AICognition's Devin can now spin up an Android Virtual Device, letting it build, run, and test Android applications directly on its own machine. The new emulator support gives Devin an Android equivalent of computer and browser use, so it can open apps, inspect behavior, reproduce issues, and verify changes. The feature is available now for teams using Devin.