Model Detail
Agents-K1
—Agents-K1 is a large language model with 4B parameters released by InternScience. The model is registered under the text-generation pipeline tag on Hugging Face, distributed under the permissive apache-2.0 license.
Agents-K1 ships with 4B parameters. Total weight footprint is approximately 4.4 GB, which is the relevant figure when planning local-inference VRAM. The apache-2.0 license is permissive, allowing commercial deployment and derivative work without per-seat fees, though attribution requirements still apply.
Agents-K1 is best fit for general-purpose chat and instruction-following workloads. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.
Agents-K1: Towards Agent-native Knowledge Orchestration
arXiv:2606.13669v3 Announce Type: replace Abstract: Current LLM-based research agents have advanced through agent orchestration, yet largely overlook scientific knowledge orchestration. Existing works often reduce papers to abstracts, surface mentions, and flat \texttt{cites} edges, omitting key ent
After the Attacker Leaves: Benchmarking Carryover Attacks on LLM Agents
arXiv:2610.28425v1 Announce Type: cross Abstract: Modern LLM agents persist state across sessions via two channels: long-lived memory files silently prepended to the system prompt, and scheduled tasks waking the agent without user input. A one-time indirect prompt injection committed to either chann
Proof-Carrying Contracts: Commitments AI Agents Cannot Break
arXiv:2610.28484v1 Announce Type: new Abstract: AI agents increasingly act on behalf of people or other agents, purchasing services, exchanging information and delegating work. What an agent must respect is written in instructions, policies or agreements, but text alone cannot make it comply. It may
TEX: Test-Time Scaling Testing Agents via Execution-based Cross-Validation
arXiv:2610.27290v1 Announce Type: new Abstract: Test-case generation is critical but remains challenging for language models. Existing test-time scaling methods either isolate candidates or aggregate traces, which do not scale to long-horizon tasks. We present TEX, a collaborative test-time scaling
FALCA: Falsification-Aware Reinforcement Learning with Hierarchical Credit Assignment for Long-Horizon Deep Search Agents
arXiv:2610.28361v1 Announce Type: new Abstract: Long-horizon deep search requires LLM agents to iteratively decompose complex information needs, gather evidence from external sources, and maintain a reliable research state under multiple constraints. A central challenge is that early mistakes can be
Self-Evolving Agents through Privileged Retrospective Distillation
arXiv:2610.27495v1 Announce Type: cross Abstract: Self-evolving agents aim to improve from their own experience rather than remain fixed after deployment, but naturally collected trajectories often lack ground-truth solutions or verifier-derived rewards. We introduce Retrospective Self-Distillation
