httpx
PulseAugur coverage of httpx — every cluster mentioning httpx across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
Wrapture Python package offers advanced monkey patching and observability
Graham Dumpleton has released wrapture, a new Python package for monkey patching that also offers capabilities for testing and observability. The tool allows for recording method calls, phased behavior, and tracing live…
-
MCP Python SDK updates from 1.x to 2.x, requiring server migration
The MCP Python SDK has been updated from version 1.x (FastMCP) to 2.x (MCPServer), requiring migration for existing Python MCP servers. This update involves renaming FastMCP to MCPServer and changes to APIs, including t…
-
Anthropic releases Python SDK v1.0.0 with httpx2 upgrade
Anthropic has released version 1.0.0 of its Python SDK, marking a significant stability milestone after numerous pre-release versions. The primary update involves replacing the HTTP client library from httpx to httpx2. …
-
FastMCP 4 migration guide details breaking changes in package structure and dependencies
The release of FastMCP 4 introduces several breaking changes that developers must address during migration. A key issue arises from the package structure change, where `pip install -U fastmcp` may not correctly install …
-
Model Context Protocol shifts to stateless architecture, introducing new security risks
The Model Context Protocol (MCP) has undergone a significant revision, moving from a stateful to a stateless architecture. This change, effective July 28, 2026, eliminates the need for session IDs and initialization han…
-
Developers cut LLM token costs with semantic caching and rate limiting
Developers are implementing caching strategies to reduce costs and improve efficiency when using free-tier Large Language Model (LLM) endpoints. One approach, SimHash, uses a hashing algorithm to identify semantically s…
-
Local LLMs face 'warm-up tax' vs. hosted endpoints
Running large language models locally can incur a "warm-up tax" where the initial request is significantly slower due to model loading times. This tax is only relevant for short-duration sessions, as longer sessions amo…
-
LLM API returns empty strings despite 200 OK, revealing a token limit bug
A developer encountered an issue where an LLM API consistently returned empty strings for summaries, despite all requests appearing successful with HTTP 200 status codes and clean JSON parsing. The root cause was traced…
-
Enforce Free LLM Token Budgets in Code, Not Discipline
Developers should treat free LLM token allowances as strict budgets and enforce them in code, rather than relying on discipline. A common pitfall is a background job consuming tokens without notice, especially during re…
-
Developer fixes LLM connection bug by managing HTTP client keep-alive
A developer encountered an intermittent connection error with a free LLM endpoint, where requests would hang for exactly thirty seconds before failing. Through detailed logging, they discovered that the issue stemmed fr…
-
Anthropic Python SDK 1.0 drops httpx, breaking mocks
Anthropic's Python SDK has released version 1.0.0, which introduces significant changes including a move from the httpx library to httpx2. This update raises the minimum Python version to 3.10 and removes several featur…
-
Free LLM Server Breaks Under Concurrent Load at 16 Parallel Requests
A performance test revealed that a free model server, identified as MonkeyCode's, broke under concurrent load at 16 parallel requests. The test, conducted using Python with asyncio and httpx, measured various metrics in…
-
Build a persistent LLM decision log with Python and MonkeyCode
This tutorial demonstrates how to build a decision log system using Python, FastAPI, and MonkeyCode's free model access. The system is designed to record not only the model's raw responses but also the actions taken bas…
-
Developer isolates LLM API empty response bug to stale TCP connections
A developer encountered persistent empty responses from an LLM API, initially suspecting the model itself. After extensive troubleshooting, the root cause was identified as a combination of a stale TCP connection, a VPN…
-
LLM server variance probes reveal unreliability of single-run tests · 2 sources tracked
Two articles explore the variance in free LLM server performance, arguing that a single run provides misleading results. The first article introduces a Python script that sends 20 requests hourly to measure latency and …
-
Developer audits LLM repeatability on free servers to distinguish model vs. server issues
A developer has created a Python script to audit the repeatability of LLM outputs from free servers, addressing the challenge of distinguishing model performance from server variability. The audit measures six signals i…
-
Simon Willison updates `llm` tool with new features and dependency fixes
Simon Willison has released updates for his `llm` tool, with version 0.33 introducing new features and version 0.32.1 addressing a dependency issue. The latest release allows for combining multiple prompt templates and …
-
Deploying LLM Gateways: Python and Node.js Tutorials Emphasize Staged Verification
Two tutorials detail the process of deploying a model gateway to a public endpoint, emphasizing a staged approach with verification at each step to avoid common deployment pitfalls. The first tutorial uses Python, FastA…
-
Developer tests free LLM server limits with Python script
A developer has created a Python script to test the performance limits of free Large Language Model (LLM) servers. The script employs a "staircase test" that gradually increases concurrency to identify when servers begi…
-
New probe.py script helps developers test LLM readiness before integration
A developer has created a Python script called probe.py to help teams evaluate new LLMs before integrating them into production. The script focuses on practical, reproducible tests for correctness, latency, and consiste…