Fetching from the wire…
01
02
03
04
05
06
07
08
Top 5 · 2026-07-13 · source-backed
Here's the number that should sit in every "agents will replace engineers" thread: 15.2%. That's the best model. The mean across 15 frontier models is 4.3%.
Tencent Hunyuan's Long-Horizon-Terminal-Bench put 15 frontier models against 46 long-horizon terminal tasks across nine categories: experiment reproduction, software engineering, multimodal analysis, interactive games, scientific computing (arXiv:2607.08964, HuggingFace Daily Papers #1 today with 41 votes). It uses dense reward grading that scores partial progress through graded subtasks instead of pass/fail. The strongest model reached 15.2% pass@1 at a 0.95 partial-reward threshold, dropping to 10.9% at a perfect 1.0. Cross-model mean: 4.3%, falling to 1.7% at perfection.
And these tasks are brutally expensive. Roughly 9.9M tokens, 231 episodes, and 85 minutes of execution each. So the frontier isn't just failing at these. It's failing at them slowly and at enormous cost.
I love this paper because it's the counterweight to everything else in this issue. Grok's cheaper. Salesforce automated half its support. Tao builds apps with agents. And then this: the moment you point an agent at a genuinely long-horizon terminal task with real state and real dependencies, it falls off a cliff. The gap between "cheap and useful for scoped work" and "autonomous on hard multi-step problems" is not closing as fast as the pricing pages suggest.
This tracks with what I hit in my own projects. Agents are excellent at bounded tasks with clear success criteria. Give them something that requires holding state across 200 steps, recovering from a failed action, and reasoning about what went wrong three moves ago, and reliability collapses. The Mosaic paper this week found the same root cause: failed actions from inaccurate state tracking under partial observability are the dominant latency and failure bottleneck, not raw model speed (arXiv:2607.09603).
What builders should do: scope aggressively. Don't hand an agent a task that needs 200 coherent steps. Decompose it into bounded units with checkable outputs, and put the state in files the next run re-reads, not in the context window. The 15% number is your reminder that "autonomous" is a spectrum, and most of the useful zone is on the short-horizon end.
Each link below shares sources, entities, or timing with this story.
Agentforce built by Salesforce / Shared entity: Salesforce / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (Agentforce built by Salesforce); both cover Salesforce; overlapping topics (action, agent, model).
Linked by a graph relationship (Agentforce built by Salesforce); both cover Salesforce; overlapping topics (action, agent).
Grok built by SpaceX / Shared entities / Shared topic
Linked by a graph relationship (Grok built by SpaceX); both cover Bench, Grok, Terminal; overlapping topics (agent, model, task).
GPT competes with Grok / Shared entities / Shared topic / Earlier coverage
Linked by a graph relationship (GPT competes with Grok); both cover Bench, Terminal; overlapping topics (frontier, model).
GPT competes with Grok / Shared entities / Shared topic / Earlier coverage / Tension
Linked by a graph relationship (GPT competes with Grok); both cover Bench, Terminal; overlapping topics (agent, frontier, model).
Agentforce built by Salesforce / Shared entity: Salesforce / Shared topic / Earlier coverage
Linked by a graph relationship (Agentforce built by Salesforce); both cover Salesforce; overlapping topics (agent, model).
Salesforce acquired Intercom / Shared entity: Salesforce / Shared topic / Earlier coverage
Linked by a graph relationship (Salesforce acquired Intercom); both cover Salesforce; overlapping topics (agent, model).
Agentforce built by Salesforce / Shared entity: Salesforce / Shared topic / Earlier coverage
Linked by a graph relationship (Agentforce built by Salesforce); both cover Salesforce; overlapping topics (agent, model).