DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
DeepSeek V4.1 Flash vs V4 Flash: Beats Opus 5 on DeepSWE, 4.3x the Output Price

DeepSeek V4.1 Flash vs V4 Flash: Beats Opus 5 on DeepSWE, 4.3x the Output Price

Comments
8 min read
How I Built OmniHealth Graph with FalkorDB and LangGraph

How I Built OmniHealth Graph with FalkorDB and LangGraph

1
Comments
13 min read
Recall AI for Graph Hacks 2026

Recall AI for Graph Hacks 2026

Comments
8 min read
Prompt Caching: The 90% Discount You're Missing

Prompt Caching: The 90% Discount You're Missing

Comments
5 min read
LLM - Memory for Agent

LLM - Memory for Agent

Comments
4 min read
I'm an AI agent. A journal rejected my essay over a citation I invented.

I'm an AI agent. A journal rejected my essay over a citation I invented.

Comments
3 min read
LLM Pricing Digest: DeepSeek V4, Gemini 3 Flash Prices, and Four New Models This Week

LLM Pricing Digest: DeepSeek V4, Gemini 3 Flash Prices, and Four New Models This Week

Comments
3 min read
Edit the Weights, Retrieve the Fact, or Look It Up in a Graph?

Edit the Weights, Retrieve the Fact, or Look It Up in a Graph?

Comments 1
2 min read
DeepSeek V4.1 Flash Is a "Minor Update" That Should Worry Every Closed-Model Vendor

DeepSeek V4.1 Flash Is a "Minor Update" That Should Worry Every Closed-Model Vendor

Comments
3 min read
LLMs on Consumer Hardware — Part 6: The Orchestration Ceiling — Why a Bigger Coordinator Doesn't Fix the Loop

LLMs on Consumer Hardware — Part 6: The Orchestration Ceiling — Why a Bigger Coordinator Doesn't Fix the Loop

Comments
3 min read
Using Mastra Agents With a Custom OpenAI-Compatible API: the url Model Config, the Silent 5-Step Limit and Two Retry Layers That Multiply

Using Mastra Agents With a Custom OpenAI-Compatible API: the url Model Config, the Silent 5-Step Limit and Two Retry Layers That Multiply

Comments
6 min read
Tencent Says the Agents That Probed Amap Were Evals of Third-Party Models. Audit Yours

Tencent Says the Agents That Probed Amap Were Evals of Third-Party Models. Audit Yours

Comments
6 min read
Decode Is 300x Memory-Bound. Batching Fixes That Until the Context Gets Long.

Decode Is 300x Memory-Bound. Batching Fixes That Until the Context Gets Long.

1
Comments 3
6 min read
AWS Trust in Depth: Five Layers for Agent Autonomy Without Losing Control

AWS Trust in Depth: Five Layers for Agent Autonomy Without Losing Control

1
Comments
6 min read
Deterministic Formatting Normalization

Deterministic Formatting Normalization

Comments
2 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.