Signal65 says AMD MI355X delivered up to 2.15x better tokens per dollar than NVIDIA B200
ryanshrout · x · 2026-07-23
Signal65’s new evaluation goes beyond raw token throughput and compares real-world inference economics on three production models.
- On blended on-demand cloud pricing, AMD Instinct MI355X delivered up to 2.15x more tokens per dollar than NVIDIA HGX B200.
- MI355X led on GPT-OSS-120B (2.15x), Qwen3-Next-80B (2.11x), and Kimi-K2.6 (1.42x).
- On Kimi-K2.6, B200 had 17% higher raw throughput, but MI355X still won on cost efficiency.
The point of the evaluation is that peak throughput and price-performance are different decisions, and both matter when sizing inference deployments.
Related event: Signal65: AMD MI355X Beats B200 in Inference Cost-Effectiveness(2 posts)→
More from Infra
- CoreWeave 5-year CDS spikes to ~1,100bps as market fears AI bubble starts as credit event — kimmonismus · 2026-10-26
- NVIDIA Plans ~$2,000 Per GPU Per Year Software Fee for Running AI Agents — LinusEkenstam · 2026-10-26
- 12 DevOps concepts to learn in order, from Linux to SRE — _jaydeepkarale · 2026-10-26
- Killed our Vector DB for SQLite BM25 and wrote a 3.6ns Rust sentinel: lessons from multi-agent loops — OpenComplaint524 · 2026-10-26
- Gemma 4 26B-A4B runs offline on a 16 GB Mac: 33/40 on TypeScript tasks at 31 tok/s — Mission-Sir-1614 · 2026-10-26
- How thousands of GPUs train one AI: a breakdown of 5D parallelism — jbhuang0604 · 2026-10-26