Topic
Running models on local or owned hardware, including framework support, memory footprint and serving cost trade-offs.
A pre-registered run reads Gemma's label logits on one 24 GB GPU and scores it against the hosted API on the same 3,880 records, where it is level on yes/no questions and 4.5 points behind on multiple choice. One temperature fitted on 50 labels closes the calibration gap.
Reality
Breeze TTS 2 scores Elo 1215 and lists at $34 per 1M characters, about a third of Eleven v3's rate. Only the inference code is Apache 2.0, and commercial use of the weights needs a written agreement from BreezeBlue.
Reality
LiteLLM works as a drop-in OpenAI replacement for teams running their own clusters, while managed gateways suit teams renting inference. OpenRouter's reported $113M round at a $1.3B valuation funds the rented side.
Reality
The Responses and Messages dialects specify tool declarations, streaming and reasoning knobs. A Java 21 library published to Maven Central at 0.18.0 argues the unspecified half is where retrieval, injection and veto actually live, and offers to host it in your own process.
Reality
AWS shows three agents sharing one AgentCore container across two hosting paths. The orchestration framework absorbs the split; the OpenTelemetry instrumentation does not.
Reality
Mixedbread's embedding model is reported to edge text-embedding-3-large by half a point. The savings case rests on an arithmetic slip and pricing that does not exist yet.
Reality
The partnership removes the external API hop for enterprises that cannot let sensitive records leave the perimeter, which also moves the inference bill off someone else's per-token pricing and onto hardware the customer has to buy and keep busy.
Reality
Alibaba scheduled a 27-billion-parameter vision-language model for August 14, alongside an already-published 2.4-trillion-parameter MoE. The smaller file is the consequential one.
Reality