gpt-oss 120B — Jalapeño vs MI355X
Head-to-head AI inference benchmark comparison of Jalapeño (OpenAI Jalapeño) and MI355X (AMD CDNA 4) on gpt-oss 120B. 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.
Setting 149 tok/s/user as the target on gpt-oss 120B, Jalapeño produces 56755 tok/s/chip ($0.01 per million tokens) and MI355X produces 15804 ($0.03). Jalapeño is 324% cheaper per token; Jalapeño delivers 259% more tok/s/chip.
At 201 tok/s/user interactivity on gpt-oss 120B, Jalapeño delivers 52968 tok/s/chip at $0.01 per million tokens; MI355X delivers 10455 tok/s/chip at $0.04. Jalapeño is 498% cheaper per token; Jalapeño delivers 407% more tok/s/chip at this point.
Jalapeño posts 47690 tok/s/chip for $0.01 per million tokens at 253 tok/s/user on gpt-oss 120B; MI355X posts 3403 tok/s/chip for $0.12. Jalapeño is 1555% cheaper per token; Jalapeño delivers 1301% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | Jalapeño:56755.3MI355X:15803.9 | Jalapeño:52967.8MI355X:10454.6 | Jalapeño:47689.7MI355X:3402.9 |
| Cost ($/M tok) | Jalapeño:$0.006MI355X:$0.026 | Jalapeño:$0.007MI355X:$0.040 | Jalapeño:$0.007MI355X:$0.122 |
| tok/s/MW | Jalapeño:50449191MI355X:7561689 | Jalapeño:47082454MI355X:5002206 | Jalapeño:42390867MI355X:1628179 |
| Concurrency | Jalapeño:~1132MI355X:~12 | Jalapeño:~662MI355X:~5 | Jalapeño:~416MI355X:~15 |