From the Bayeux tapestry to Glastonbury: top tips and tricks to grab the hottest tickets36m ago Protected: DiVeR: Decision-Critical Verifier Learning for VLA Test-Time Scaling43m ago Meta, OpenAI and Uber Just Taught AI Agents to Talk First. What About When to Stay Quiet?1h ago IBM allows on-prem deployment of its Bob agentic development platform1h ago Industrial robots working in factories surpass 5M worldwide2h ago Your boss might be bluffing about how much AI knowledge they really have2h ago OpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a day3h ago Prime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models3h ago Anthropic turns to Hindu philosophy to teach Claude right from wrong3h ago Microsoft AI Releases MAI-Transcribe-2-Streaming: #1 Real-Time Speech-to-Text Model on Artificial Analysis3h ago AI's race to transform the world before the money runs out3h ago AI breaks India’s linear growth model, but campus hiring remains weak: Nasscom’s Rajesh Nambiar3h ago America’s data center fight is a preview of what's to come for the rest of the world3h ago Apple to Tighten Mac Data Controls in Guard Against AI Agents3h ago

From the Bayeux tapestry to Glastonbury: top tips and tricks to grab the hottest tickets36m ago Protected: DiVeR: Decision-Critical Verifier Learning for VLA Test-Time Scaling43m ago Meta, OpenAI and Uber Just Taught AI Agents to Talk First. What About When to Stay Quiet?1h ago IBM allows on-prem deployment of its Bob agentic development platform1h ago Industrial robots working in factories surpass 5M worldwide2h ago Your boss might be bluffing about how much AI knowledge they really have2h ago OpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a day3h ago Prime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models3h ago Anthropic turns to Hindu philosophy to teach Claude right from wrong3h ago Microsoft AI Releases MAI-Transcribe-2-Streaming: #1 Real-Time Speech-to-Text Model on Artificial Analysis3h ago AI's race to transform the world before the money runs out3h ago AI breaks India’s linear growth model, but campus hiring remains weak: Nasscom’s Rajesh Nambiar3h ago America’s data center fight is a preview of what's to come for the rest of the world3h ago Apple to Tighten Mac Data Controls in Guard Against AI Agents3h ago

Agents Machine Learning11 min reading time

How to Build a Model Router in the Harness

LangChainPublished 19 hours ago

Read the full article

LangChain developed a model router for its Open SWE coding agent that selects appropriate AI models per task, reducing costs by 64% without quality loss. The router leverages task context within the agent harness to optimize model use, avoiding expensive frontier models for simpler tasks.

#agentic-software-development #anthropic #claude-haiku-4-5 #langchain #open-swe

More in Agents

Agents6 min read

The Battle to Be Your Personal AI Agent Is Here

Covered by 5 sources3d ago

Agents4 min read

IBM allows on-prem deployment of its Bob agentic development platform

Covered by 2 sources2d ago

Meta, OpenAI and Uber Just Taught AI Agents to Talk First. What About When to Stay Quiet?

MarkTechPost1h ago