AI coding

Anthropic released Claude Opus 5.5 on OpenRouter, the first model in the Claude 5.5 series. It beats Opus 5 and Fable 5.1 on agentic coding, knowledge work, and computer use, with a 1M context window. Pricing is $4 per million input tokens and $20 per million output tokens, 20% cheaper than Opus 5. The post doesn't include benchmark scores or latency figures.

Claude Opus 5.5 is the first model in the Claude 5.5 family. It matches Claude Fable 5.1 on most tasks and costs 40% less to run than Opus 5. Claude Devs adds it's ~30% faster per task. Claude Code's 5-hour session limit increased 20% today; lower pricing means 25% more usage within the cap. Pro, Max, and Team users also get a one-time quota reset. Terminal-Bench 4.0 scores lead across effort tiers.

Anthropic dropped Claude Opus 5.5, the first model in the 5.5 family. It matches Fable 5.1 on most tasks and costs 40% less to run than Opus 5. The author notes clearer communication, better token efficiency, and availability across all effort levels. The 5-hour rate limit is raised and a banked reset feature is added. The post doesn't disclose specific benchmarks or pricing.

Anthropic released Opus 5.5, the first model in the Claude 5.5 family. The company says it matches Claude Fable 5.1 on most tasks, costs 40% less to run than Opus 5, and has lower per-token pricing with more efficient token usage. It supports all effort levels and is already available in Claude Code. The post doesn't disclose exact pricing or benchmark comparisons.

Anthropic launched Opus 5.5 on Tuesday, calling it “the strongest-performing model we've tested to date.” The company claims new state-of-the-art results in coding and knowledge work, with lower prices than previous Opus models. The post doesn't disclose specific pricing, benchmark scores, or a direct comparison with Fable, so I'd hold off on the “strongest” claim until third-party evals land.

JetBrains consolidates six months of agentic development experiments into Air, an open system spanning developers, teams, and orgs. It goes beyond the IDE with multi-surface, multi-service design, betting on a multi-vendor future. The post confirms Central CLI, shared context, cloud agents, automations, and AI cost controls are already rolling out, but pricing and GA dates aren't disclosed.

Xiaomi released the MiMo-V2.6 series. The Pro version ranks #1 among open weights models on the Artificial Analysis Intelligence Index with a score of 46, at a training cost of $3M. A Flash variant targets efficiency, and an UltraSpeed variant offers 20x faster output. The technical report details RL scaling across three axes: larger batches and throughput, richer multi-task environments, and more grader compute. Code and training recipes are open-sourced, but the 7k+ task datasets are not yet released. Former DeepSeek engineer Fuli Luo, now at Xiaomi, previously live-streamed the training runs.

ZCode was caught silently packaging entire Git histories for cloud upload, with .git objects making up 86.6% of snapshots. HarnessTax benchmarked Claude Fable 5 across frameworks: Claude Code and minimal Pi achieved near-identical success rates but a 2x cost gap. A Microsoft architect replaced multi-model agent loops with a single model reading skill docs and calling tools directly—halving API calls but increasing total tokens by 22%. Each story unpacks one number and a reminder to check which layer a metric actually measures.

NVIDIA positions Nemotron 3.5 Lightning as the execution layer in agent workflows—handling frequent tool calls, file reads, and result checks rather than heavy planning. It's a 30B MoE model that activates only ~3B parameters per token, keeping latency and cost low for high-volume calls. It complements, not replaces, Nemotron 3 Ultra. Weights are open, with tool calling and structured output support. Context goes up to 1M tokens, though OpenRouter's standard tier caps at 262K. Worth a look if your agent makes many model calls per run.

Xiaomi released two omni-modal models: MiMo-V2.6-Pro and Flash. The Pro version scored 1628 on Code Arena WebDev, up 153 points from MiMo-V2.5-Pro's 1475, landing around 10th overall and ~3rd among open-weight models under MIT license. The post doesn't disclose Flash's benchmark numbers or parameter counts.

Xiaomi MiMo dropped two fully multimodal open-source models. The Pro version matches Claude Opus 5 and GPT-5.6 Sol on most agent benchmarks and scores 46 on the Artificial Analysis Intelligence Index—the highest among open-source models so far. Capabilities span coding, computer use, 3D reasoning, and creative tasks. The post doesn't disclose parameter counts, training details, or where Flash sits in the lineup, so I'd hold off on direct comparisons for now.

Grok 4.7 uses a larger base model and a longer RL run on harder, multi-hour tasks. It scores 46.3% on CursorBench 4.0, ahead of GPT-5.6 Sol Max (41.7%) but behind Fable 5.1 Max (51.8%). Pricing stays at $2/$6 per million input/output tokens, same as Grok 4.6, with double the speed. Safety stack is new: only 3.3% of risky cyber prompts get through, and it hits 62.4% on LatchBio's biosafety benchmark. Available today in Cursor, Grok Build, and the API.

Will Larson pushed his agent setup at Imprint further: give an agent a Linear project, and it first checks for a Notion RFC and Datadog/Snowflake dashboards—prompting you to create them if missing. It then scans task status, adds newly identified work, and picks up unblocked tasks to write PRs, nudge reviews, or ask clarifying questions. After a task completes, if the project description is stale, it re-runs the full loop. Larson says this forced him to hand over goal-state he used to hoard, so agents can now judge direction. He plans to move this loop from local to the company's internal 'Agent Fleet' orchestrator. The post does not disclose performance numbers or cost.

A Microsoft team used an in-house AI agent system called Nachete to rewrite the Copilot runtime from C# to Rust, at a total cost of about $120K. The agents ran 7 iterations—writing code, compiling, and fixing errors—producing 125K lines of Rust that compiled on the first try. Humans only did code review and security audit. The team estimates a manual rewrite would have cost $1.1M and taken 9 months, though the post doesn't detail how that baseline was calculated.

This post describes a pattern for running long Claude Code sessions reliably: separate coordination from execution. One long-lived session assigns work, verifies claims, and records lessons, while short-lived sessions do the actual coding. State lives in a durable external board, not in context. The key discipline is to never trust an agent's self-report—re-run the commands and check exit codes. cmux is used to spawn execution workspaces. The pattern is essentially orchestrator-worker; the author calls it Chief of Staff but notes it's different from Anthropic's calendar-managing agent of the same name.

StepFun introduces Step 5 Preview, a 600B-parameter MoE model with 27B active per token, a 1M-token context window, and vision support. It scores 67.7 on DeepSWE v1.1, ahead of Kimi K3 and GLM-5.3 but behind GPT-6 Astra and Claude Opus 5. On the in-house StepCodeBench it hits 49.0, again leading domestic models and trailing the two US labs. On FrontierFinance it reaches 66.4, second only to Claude Opus 5. Artificial Analysis gives it an intelligence index of 44; StepFun claims substantially lower cost per task at comparable intelligence. The post does not disclose API pricing, release timeline, or training details.

Anthropic's test-impact-analysis service saw 25× load growth in six months. Three patches bought 70 days, 29 days, then less than a day of stability. One engineer rewrote it in three weeks—a task the author estimates would have taken a quarter a year ago. The rewrite cost dropped while the hidden cost of patching rose, shifting the break-even point earlier. The post does not disclose the new system's exact running cost, defect rates, or production incident data.

Diogo Almeida, a former OpenAI researcher who co-invented RLHF, built a model called Jev that optimizes for code execution rather than human language. It's still a transformer, but TypeSafe AI designed it to run programs and call APIs directly, acting more like an automation agent. Developer reception has been enthusiastic. The post doesn't disclose benchmark scores, pricing, or whether weights will be open.

Dan Luu notes a growing trend of developers blindly trusting LLM outputs and acting as a 'meat proxy' in a loop. By September 2026, this brain-off approach can produce barely functional software, but Luu argues that if LLMs get good enough to work unsupervised, companies will just run the loop themselves and lay off the human. He shares concrete failures, including an AI bot weaker than a simple heuristic bot and a commercial product trapping users in an infinite loop. Luke Burton adds that high-value tasks still require constant supervision due to too many unknown unknowns.

Justin Cormack built an S3-compatible storage system with AI, reaching 350k lines of Rust. He ran 1,500 tests against real S3 as an oracle, which caught real S3 500 errors. Chasing 100% coverage backfired—agents wrote trivial tests. Docs were often wrong, and AI was bad at finding edge cases from them. His hard rule: fix flaky tests immediately, or the agent learns to ignore failures.