GLM 5/5.1 · GB300 NVL72 · Precision comparison

GB300 NVL72: FP4 vs FP8 precision comparison

How FP4 and FP8 precision affect GLM 5/5.1 inference on GB300 NVL72 (NVIDIA Blackwell). Throughput, latency, and cost across LLM workloads. Use the chart controls below to switch sequences and metrics — same interactions as the main inference chart.

Near the low end of the 19–115 tok/s/user interactivity band, at 43 tok/s/user on GLM 5/5.1 (GB300 NVL72): FP4 runs 10,120 tok/s/chip at $0.063 per million tokens, FP8 runs 2,618 at $0.25. FP4 delivers 3.9× the tok/s/chip on the same chip, which makes it 74% cheaper per token. Precision changes affect both inference speed and model quality — consult the evaluation tab for accuracy benchmarks.

At 67 tok/s/user on GLM 5/5.1 (GB300 NVL72), FP4 delivers 6,960 tok/s/chip at $0.092 per million tokens; FP8 delivers 479 tok/s/chip at $1.34. FP4 delivers 15× the tok/s/chip on the same chip, which makes it 93% cheaper per token. Lower-precision quantization trades model accuracy for throughput — check the evaluation page for quality impact.

FP4 posts 3,619 tok/s/chip for $0.18 per million tokens at 91 tok/s/user on GLM 5/5.1 (GB300 NVL72); FP8 posts 171 tok/s/chip for $3.74. FP4 delivers 21× the tok/s/chip on the same chip, which makes it 95% cheaper per token. Quantization-level accuracy differences are tracked on the evaluation tab. (Numbers reflect the default 1k/1k selection for this URL — table and chart below update if you change sequence or model in the controls. Each side uses the best available serving configuration for that precision, which may include speculative decoding such as MTP where recipes exist — the same convention as the other comparison pages.)

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)
FP410,120FP82,618
FP46,960FP8479
FP43,619FP8171
Cost ($/M tok)
FP4$0.063FP8$0.25
FP4$0.092FP8$1.34
FP4$0.18FP8$3.74
tok/s/MW
FP44,773,597FP81,235,080
FP43,283,232FP8225,958
FP41,706,884FP880,846
Concurrency
FP4~3,606FP8~3,177
FP4~1,310FP8~307
FP4~567FP8~79
GLM 5/5.1: FP4 versus FP8 throughput and cost at matched interactivity levels on GB300 NVL72
FP4 versus FP8 throughput and cost per million tokens on GB300 NVL72 for this comparison's canonical default workload
Open full-size image ↗