DiggTop ⌕

Radeon AI Pro R9700 x3 vs Tesla P100 x2

Pick a different pair →

Best sourced output tok/s per model (single-stream generation) — engine and version, quantization, context and a source link on every row. Hover a value for the quant.

Sourced data on this pair is thin — most of the board is still partial. Submit your own run and help fill the matrix: one prompt, three steps, no downloads.
Radeon AI Pro R9700 x39 sourced runs · best 223.4 tok/s
Tesla P100 x20 verified runs · best — tok/s

Specs and evidence

Radeon AI Pro R9700 x3Tesla P100 x2
VRAM32 GB32 GB
Sourced runs on the board90
Models measured90
Best output tok/s223.4 tok/s—
Highest measured peak VRAM25.8 GB—
Street price (latest snapshot)——

Sourced rows only. Peak VRAM is recorded by the person who ran the benchmark, so it is a floor for what the card can hold, not a specification. Price snapshots.

The measured matrix

ModelRadeon AI Pro R9700 x3Tesla P100 x2
gpt-oss-120b223.4 MXFP4—
Nemotron-Cascade-2-30B-A3B95.4 Q8_0—
Qwen3-VL-8B-Instruct58.8 Q8_0—
Muse-Glimmer-30B47.8 UD-Q8_K_XL—
Ling-3.0-flash39.3 Q4_0_ROCMFP4_STRIX_LEAN—
Ornith-1.5-35B-A3B35.4 BF16—
Ornith-1.0-9B31.2 BF16—
Qwen3.8-Flash-Next28.1 UD-IQ3_XXS—
gemma-4-31B-it23.9 Q8_0—

9 rows · 9 models on Radeon AI Pro R9700 x3, 0 on Tesla P100 x2 — no verified overlap yet, each side shows its own bests · data as of Oct 17, 2026. Full rules: methodology.

Related reading: RTX 5090 vs 3090 · quantization · Value index · Full board

↑