AI Coding Live News

Independent evaluator Artificial Analysis published Tencent Hunyuan Hy4 results on the Intelligence Index v5 on November 5 (08:28 UTC): a composite score of 35, tying Grok 4.7 (xhigh) and one point behind Claude Haiku 5.5 (max). Among China-based labs it ranks second only to Kimi K3.1 Pro (46). On agentic coding it scores 31% on FrontierCode 2.0 and 37% on Terminal-Bench 4.0; cost per Intelligence Index task is about $0.78, roughly one-tenth of same-score Grok 4.7.

Zoom

Hugging Face, BitRobot, the Singapore Embodied AI Centre, NUS and others launched RoboColosseum on November 4, a distributed real-robot evaluation framework covering four embodiments: SO-101, Franka, Bimanual-YAM and the Unitree G1 humanoid. The first study ran 5,699 blinded pairwise comparisons at 124 sites across 5 VLA policies and 3 frontier models. Claude Opus 5.5 and GPT-6 Astra beat their VLA counterparts on average task progress by 12.1 and 11.0 points, while Grok 4.7 trailed by 7.5. All rollouts are open on Hugging Face, and the project pledges 10,000 free evaluation hours to academic labs.

Zoom

Cognition, the maker of Devin, released FrontierCode 2.0 on Oct 27. It keeps the method of maintainer-written tasks graded on whether the code is actually mergeable, but draws tasks from more repositories, adds security patches and performance work, uses deliberately underspecified prompts and adds many thoroughness criteria. In Cognition's results Claude Opus 5.5 leads at 45.5%, GPT-6 Astra follows at 42.8%, and Grok 4.7 and Gemini 4 Argon drop into the 20s, leaving the benchmark far from saturated.

Zoom

On October 21 Letta released FrontierMemory 1.0. Instead of quizzing agents on facts such as birthdays or hobbies, it gives an agent a set of past trajectories, has it write memory or skills, and scores that memory by how many coding and knowledge-work tasks a task agent then solves. Version 1.0 has 15 tasks adapted from public benchmarks such as DeepSWE, SkillsBench and APEX-Agents. On the model board (memory written in Letta Dream, tasks run in Letta Code) Opus 5.5 and GLM-5.3 both score 73.3% and Grok 4.7 scores 33.3%; on the harness board Letta's own Letta Dream ranks first. All numbers are Letta's own.

Zoom

On 2026-10-02 SpaceXAI said Grok 4.7 is 50% off on Ramp Router through October 6. xAI list price remains $2/$6 ($4/$12 for the whole request at ≥200k prompt tokens), and the Router catalog still shows $2.00/$6.00. Routing is free through 2026 with $26 in new-user credits. The discount is not in the docs; check request cost logs before treating $1/$3 as real.

Zoom

OpenClaw shipped v2026.9.6 on 2026-09-23 (macOS DMG rebuilt 2026-09-24 after launch crash #156861): clearer managed-update outcomes, restart recovery for unfinished work, complete 30-day Usage reporting, a GitHub reader, remote workspace Files/Memory/Skills, live meeting notes, optional TypeSafe Jev Decision Models, and chat models Claude Opus 5.5, GPT-6 Sol/Luna, and Grok 4.7. Scale: 2,614 PRs, 178 direct commits, ~350 contributors.

Zoom

Elon Musk posted Grok 4.7, saying the model is doing real engineering work at Tesla and that results are strongest with xAI’s Build harness. It is available in Cursor, Grok Build, and via API.

Zoom

@XFreeze reports Grok 4.7 Fast is live in Grok Build and Cursor: the same Grok 4.7 model on faster infra with roughly 2× output speed, at about 2× standard token rates. It is not in Grok Build’s free tier and is not on the public xAI API. Screenshots show Cursor’s “Grok 4.7 High Fast” picker and Grok Build’s Fast variant note.

Zoom

Elon Musk posted “Grok 4.7.” Same-day discussion said the model improves on Grok 4.6 for agentic coding and knowledge work, and that it works best with the xAI Build harness.

Zoom

Elon Musk announced Grok 4.7. He described it as a strong mix of intelligence, speed, and low cost, said it works especially well with xAI's Build harness, and positioned it as a competitive everyday workhorse for agentic coding.

Zoom

Elon Musk posted an announcement of Grok 4.7 with an image, signaling the latest Grok model from xAI.

Zoom

@ArtificialAnlys published AA-Briefcase results: Grok 4.7 sits behind only Anthropic models, ranking just after Opus 5 at about half the cost per task. Versus Grok 4.6 on public AA-Briefcase-Lite due-diligence decks, Analytical Quality Elo rose 1698 to 1994 while Presentation Elo slipped 1531 to 1499. Example deck API cost: Grok 4.7 (xhigh) about $8 vs Grok 4.6 (xhigh) about $4.40.

Zoom