Qwen 3.8 Flash Next 176B-A6B · Chip comparison

Qwen 3.8 Flash Next 176B-A6B — GB300 NVL72 vs MI355X

Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on Qwen 3.8 Flash Next 176B-A6B. 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.

AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX →

Near the low end of the 35–389 tok/s/user interactivity band, at 123 tok/s/user on Qwen 3.8 Flash Next 176B-A6B: GB300 NVL72 runs 187,009 tok/s/chip at $0.003 per million tokens, MI355X runs 45,057 at $0.009. GB300 NVL72 is 63% cheaper per token; GB300 NVL72 delivers 4.2× the tok/s/chip.

Setting 212 tok/s/user as the target on Qwen 3.8 Flash Next 176B-A6B, GB300 NVL72 produces 117,189 tok/s/chip ($0.005 per million tokens) and MI355X produces 28,471 ($0.015). GB300 NVL72 is 63% cheaper per token; GB300 NVL72 delivers 4.1× the tok/s/chip.

At 300 tok/s/user interactivity on Qwen 3.8 Flash Next 176B-A6B, GB300 NVL72 delivers 77,411 tok/s/chip at $0.008 per million tokens; MI355X delivers 21,543 tok/s/chip at $0.019. GB300 NVL72 is 57% cheaper per token; GB300 NVL72 delivers 3.6× the tok/s/chip at this point. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)

View performance per dollar →

Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.

Metric
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Throughput (tok/s/chip)
GB300 NVL72187,009MI355X45,057
GB300 NVL72117,189MI355X28,471
GB300 NVL7277,411MI355X21,543
Cost ($/M tok)
GB300 NVL72$0.003MI355X$0.009
GB300 NVL72$0.005MI355X$0.015
GB300 NVL72$0.008MI355X$0.019
tok/s/MW
GB300 NVL7288,211,749MI355X21,558,168
GB300 NVL7255,277,807MI355X13,622,634
GB300 NVL7236,514,834MI355X10,307,840
Concurrency
GB300 NVL72~41MI355X~9
GB300 NVL72~21MI355X~5
GB300 NVL72~13MI355X~2