Qwen 3.5 397B-A17B — MI300X vs MI355X
Head-to-head AI inference benchmark comparison of MI300X (AMD CDNA 3) and MI355X (AMD CDNA 4) on Qwen 3.5 397B-A17B. 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.
MI300X / MI355X on Qwen 3.5 397B-A17B at 43 tok/s/user: 346 / 2808 tok/s/chip, $0.76 / $0.15 per million tokens. MI355X is 410% cheaper per token; MI355X delivers 712% more tok/s/chip.
Around the middle of the 34–71 tok/s/user interactivity band, at 53 tok/s/user on Qwen 3.5 397B-A17B: MI300X runs 190 tok/s/chip at $1.38/M tokens, MI355X runs 2454 at $0.17/M. MI355X is 720% cheaper per token; MI355X delivers 1195% more tok/s/chip.
Setting 62 tok/s/user as the target on Qwen 3.5 397B-A17B, MI300X produces 123 tok/s/chip ($2.13 per million tokens) and MI355X produces 2189 ($0.19). MI355X is 1026% cheaper per token; MI355X delivers 1684% more tok/s/chip. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MI300X:345.8MI355X:2808.3 | MI300X:189.5MI355X:2453.8 | MI300X:122.7MI355X:2189.4 |
| Cost ($/M tok) | MI300X:$0.758MI355X:$0.148 | MI300X:$1.385MI355X:$0.169 | MI300X:$2.135MI355X:$0.190 |
| tok/s/MW | MI300X:248748MI355X:1343685 | MI300X:136357MI355X:1174060 | MI300X:88306MI355X:1047577 |
| Concurrency | MI300X:~34MI355X:~132 | MI300X:~15MI355X:~90 | MI300X:~8MI355X:~71 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.