Kimi K2.5/K2.6/K2.7-Code 1T — Jalapeño vs MI355X
Head-to-head AI inference benchmark comparison of Jalapeño (OpenAI Jalapeño) and MI355X (AMD CDNA 4) on Kimi K2.5/K2.6/K2.7-Code 1T. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
Jalapeño posts 12110 tok/s/chip for $0.03 per million tokens at 76 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T; MI355X posts 2213 tok/s/chip for $0.19. Jalapeño is 546% cheaper per token; Jalapeño delivers 447% more tok/s/chip.
Throughput at 92 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T: Jalapeño hits 11370 tok/s/chip, MI355X hits 1821. Per-million costs land at $0.03 and $0.23 respectively. Jalapeño is 637% cheaper per token; Jalapeño delivers 524% more tok/s/chip.
Jalapeño / MI355X on Kimi K2.5/K2.6/K2.7-Code 1T at 107 tok/s/user: 10575 / 1418 tok/s/chip, $0.03 / $0.29 per million tokens. Jalapeño is 781% cheaper per token; Jalapeño delivers 646% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | Jalapeño:12110.3MI355X:2213.5 | Jalapeño:11369.9MI355X:1821.0 | Jalapeño:10575.4MI355X:1418.1 |
| Cost ($/M tok) | Jalapeño:$0.029MI355X:$0.188 | Jalapeño:$0.031MI355X:$0.229 | Jalapeño:$0.033MI355X:$0.294 |
| tok/s/MW | Jalapeño:10764719MI355X:1059089 | Jalapeño:10106603MI355X:871276 | Jalapeño:9400353MI355X:678507 |
| Concurrency | Jalapeño:~1500MI355X:~13 | Jalapeño:~1037MI355X:~9 | Jalapeño:~756MI355X:~5 |