SemiAnalysis releases TPUv7 agentic inference benchmark
Understand this story
SemiAnalysis has published the first long-context, multi-turn agentic inference benchmark results for TPUv7 on InferenceX/AgentX, covering four models: Muse Spark Open, GLM-5.2, Qwen3.5 397B, and Kimi K3. SemiAnalysis says the results show that, for external customers, TPUv7's performance per unit cost is either comparable to NVIDIA GPUs or reaches best-in-class levels. This is the organization's first official benchmark release on TPUv7's agentic inference capabilities.
AI overview based on source reports · Updated 10 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- SemiAnalysisThe Iron(wood) Throne: Google's TPU Stakes Its Claim on Inference
SemiAnalysis reports first official long-context, multi-turn agentic inference benchmark results on TPUv7 (Ironwood) across four models on InferenceX/AgentX, finding TPUv7 either competitive or best-in-class on performance per dollar for external customers, including 1.2x to 2.7x the tokens per dollar of B200, B300 and MI355X on GLM-5.2 FP8. The results came through ecosystem partners including SGLang (RadixArk), vLLM (Inferact) and llm-d, and the piece argues NVIDIA's bureaucracy and stingy cluster contributions to open-source maintainers are among the biggest risks to its moat. It also notes NVIDIA's GB300 FP4 counter submission still trails TPUv7 FP8 on end-to-end performance per dollar, while FP4 GPUs retain a per-dollar lead on GLM-5.3 until TPUv8 adds FP4 support, expected in Q1 2027.
Activity for this story
- Current comparable activity
- 8
- Peak comparable activity
- 10Nov 6 04:00
- Change in the past 24 hours
- –
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.