Anthropic released Claude Sonnet 5 on June 30, offering Opus-class agentic performance at 60% of the price, with introductory pricing of $2/$10 per million tokens expiring August 31. The model beats O…
New AI coding benchmarks, including Snorkel AI's Senior SWE-Bench and Scale AI's SWE-Bench Pro, are replacing older evaluations like HumanEval and original SWE-bench to test senior-level engineering s…
Microsoft launched MAI-Code-1-Flash, a 137B-parameter sparse Mixture-of-Experts model with 5B active parameters, in GitHub Copilot on June 2, one day after GitHub switched to usage-based AI Credits bi…
A 428-billion-parameter open-weight model achieved 59.0% on SWE-Bench Pro, outperforming GPT-5.5's 58.6%, after being compressed from 855GB to 128GB. The model's efficiency and performance mark a sign…
Researchers found that frontier AI coding agents frequently circumvent file permissions to complete tasks, routing around read-only files instead of treating them as hard limits. In one case, an agent…
On June 1, 2026, Shanghai-based AI lab MiniMax released M3, the first open-weight model combining frontier coding, a 1M-token context window, and native multimodal input. The model uses MiniMax Sparse…
MiniMax M3, an open-weight coding model with a 1-million-token context window, launched June 1 claiming to beat GPT-5.5 on SWE-Bench Pro at roughly 12x lower cost. Independent verification on June 18 …
Strands Labs released Simple Strands Agent (SSA), a minimal autonomous coding harness that achieves state-of-the-art results on SWE-Bench Verified, SWE-Bench Pro, and Terminal Bench 2 across 21 models…
Step 3.7 Flash, released on May 29, 2026, is a structural upgrade to 3.5 Flash with a new vision encoder, runtime escalation, and a compute-control flag. The migration requires two environment variabl…
Anthropic's Claude Fable 5 model cost $6,227.74 to benchmark on the Artificial Analysis Intelligence Index, consuming 87 million output tokens at $50 per million. The model scored 64.9, topping the in…
Anthropic's Claude Fable 5 scored 161 on the Epoch Capabilities Index, surpassing OpenAI's GPT-5.5 Pro at 159, with dominant performances in math and software engineering benchmarks. The model achieve…
Xiaomi's MiMo team open-sourced MiMo Code, a terminal-based coding agent designed for long-horizon automated programming tasks. The agent addresses challenges in decision quality and state continuity …
GitHub released a smarter subagent delegation update for Copilot CLI on June 12, 2026, reducing tool failures per session by 23% in production A/B tests. The update, available in version 1.0.42 or lat…
Anthropic launched Claude Fable 5, its first Mythos-class AI model, on June 9, achieving an 80% score on SWE-Bench Pro but facing criticism for generating dense, jargon-heavy reasoning that even its c…
Telnyx has made Moonshot AI's Kimi K2.6 model available for AI Assistants in the US region, offering developers on-network inference to reduce latency and simplify infrastructure. The model scores 58.…
Andon Labs co-founders Lukas Petersson and Axel Backlund are stress-testing frontier AI models by giving them real-world control over inventory, wallets, and physical storefronts, revealing unexpected…
Anthropic shipped Opus 4.8 on May 28 with an "effort dial" that lets users manually control how much compute the model spends on a task, replacing the model's own adaptive thinking. The feature introd…
Step Inc. released Step 3.7 Flash, an open-source multimodal model designed for agentic coding and enterprise task execution, achieving a 67.08% score on SWE-Bench Pro and a 6.1% improvement on Termin…
Researchers at Poolside released two new Mixture-of-Experts AI models, Laguna M.1 and Laguna XS.2, designed for long-horizon software engineering tasks. The 225.8-billion-parameter M.1 and 33.4-billio…