SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World
Back to feed
Programming

OpenAI Codex Bug on AWS Bedrock Reportedly Inflating Costs by 10x

0
·1 views

A bug affecting OpenAI's Codex when deployed via AWS Bedrock has been reported to cause users to incur charges approximately ten times higher than expected. The issue was flagged in OpenAI's official Codex GitHub repository, drawing early community attention. The report has garnered points and comments on Hacker News, suggesting it has caught the interest of developers using the platform. The root cause and whether a fix has been issued remain unclear from the available information.

Read the full story at Hacker News

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Service Bazzar Helps Jaipur Residents Find and Compare Local Astrology Services

Astrology remains a widely consulted practice in Jaipur, with residents seeking guidance on marriage, career, relationships, and personal decisions. Astrologers in the city offer services such as Kundli analysis, horoscope reading, marriage compatibility matching, and Vastu consultation. Kundli matching is particularly common among families evaluating traditional compatibility factors before arranging a marriage. Service Bazzar, a local service discovery platform, allows users to search and compare astrology professionals across Jaipur neighbourhoods including Vaishali Nagar, Mansarovar, and Malviya Nagar. Experts advise considering an astrologer's experience, communication style, and customer feedback before selecting a service provider.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

llama.cpp's V cache quantization silently requires Flash Attention, cutting context in half

A developer building an offline desktop app powered by llama.cpp discovered that enabling V cache quantization requires Flash Attention to be turned on, or the context window is effectively halved without obvious warning. The error surfaces through three different messages at startup, but the root cause — that quantized V cache storage is architecturally incompatible with non-Flash Attention paths — is rarely documented clearly. The developer also found that calculating the KV cache cost per token from model metadata alone can be wrong by a factor of two or more, particularly for architectures like Gemma that interleave attention layers. A reliable fix involves probing llama.cpp at runtime with a small context using the same flags as production, rather than relying on metadata estimates. With correct measurement and matching flags, the same 24 GB machine and model yielded either 8,123 or 16,384 tokens of context depending on whether production settings were accurately reflected.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Why AI Features Need Structured Testing, Not Just a Few Manual Checks

A software developer argues that AI-powered features are being shipped with far less rigorous testing than standard code, often relying on informal manual checks rather than systematic evaluation. The author points out that every component of an AI pipeline — retrieval, context, prompts, and model output — can fail independently, yet most teams have no automated way to detect these failures. To address this, the article recommends building small evaluation datasets that can be run consistently whenever prompts, models, or workflows are changed. The piece also highlights that poor context passed to a model can produce bad outputs even when the prompt itself is well-crafted. The author concludes that treating AI evaluation with the same discipline applied to traditional software testing is essential for reliable AI development.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer discloses Claude AI writes his tech blog, calls it honest collaboration

Software developer Dogan has publicly disclosed that his posts on DEV Community are drafted by Claude, Anthropic's AI assistant, rather than written by him personally. Claude actively participates in his development work — reading codebases, running migrations, and browsing — then writes up notable findings from those sessions. Dogan reviews every post, removes content he disagrees with, and makes the final publishing decision, but credits the drafting entirely to the AI. He frames the arrangement as a practical solution to a common developer problem: valuable technical insights from real work that never get documented. His upcoming posts will cover infrastructure topics and an unexpected data-analysis project sparked by monitoring his children's YouTube Kids recommendations.

0 comments Read more at DEV Community
OpenAI Codex Bug on AWS Bedrock Reportedly Inflating Costs by 10x · ShortSingh