Topic
A category of AI models tuned for multi-step coding tasks—bug finding, patching, test running—evaluated on agentic benchmarks like Terminal-Bench.
The MIT-licensed 320B model card claims it beats GLM-5.2 at a tenth of the price and approaches Claude Opus 4.8 on coding, but it names no dollar rate, and the comparisons are largely the vendor's own.
Reality
A 61 on Artificial Analysis's index no longer requires flagship pricing. That changes what agent workloads should cost this quarter. The same release also grew dearer than its own predecessor and slipped on two evaluations.
Reality
The release attaches speculative decoding to weights it publishes under MIT, which lowers what a self-hosting threat costs to stand up, even though the performance claim behind it is still the vendor's own.
Reality
Two weeks after release, Grok sits in Microsoft's, Amazon's and Google's catalogs. The Foundry listing is a public preview, which is not what the Bedrock deployment got.
Reality
Seven days after launch, xAI's flagship sits inside AWS procurement with a 500K context and four reasoning tiers. The rate card is flat; the effort dial is where the cost moves.
Reality
Z.ai's August 14 post claims post-training gains for coding agents, but the company's release notes still stop at GLM-5.1 and there is no API endpoint, model identifier or weight download.
Reality
Z.ai says all of GLM-5.3's coding gains came from post-training on tenfold more long-horizon task environments. The uneven benchmark jumps tell you where that money actually landed.
Reality