Compare

Loading community benchmarks…

Community benchmarks

Compare

Put two chips, runtimes, or runtime versions side by side. Everything else is held fixed or called out, so a ratio is worth exactly what the matched runs say.

Add your benchmark

Runs compared: 9
Like-for-like pair: 1
Chips available: 6

Compare

Side A

Side B

Model

Runtime

Runtime version

Quantisation

Backend

Prefill size

Decode size

Cooldown

6 runs vs 3 runs · 1 like-for-like pair

Held fixedModel: Nemotron 3 Nano 30B A3BPrefill: PP512Decode: TG128

A side mixes quantisations. Pick a quantisation to compare the same weights.
The sides differ in cooldown, conditioning, decode workload. Ratios may reflect benchmark protocol rather than the chip.

Chip comparison

Apple M5 Max vs Apple M1 Max

Ratios read side A relative to side B. Geometric mean over 1 like-for-like pair.

Decode: 1.62×+62.4%
Prefill: 5.68×+468%

Side A

Apple M5 Max

6 runs

2 contributors

Latest 11 Sep 2026

Decode · TG128: 179.2tok/s median
best 184.4 · worst 109.2
Prefill · PP512: 4,942tok/s median
best 4,984 · worst 4,527

Models: NVIDIA-Nemotron-3-Nano-30B-A3B
Runtimes: BaseRT
Versions: BaseRT 0.2.4
Quantisations: Q4
Backends: Metal
Cooldown: off, on
Conditioning: idle_reset_then_warm, warmup_only
Decode workload: TG128
Harness schema: basert-benchmark-harness/1

Side B

Apple M1 Max

3 runs

1 contributor

Latest 20 Sep 2026

Decode · TG128: 61.1tok/s median
best 110.3 · worst 60.6
Prefill · PP512: 858tok/s median
best 870 · worst 857

Models: NVIDIA-Nemotron-3-Nano-30B-A3B-Q4, NVIDIA-Nemotron-3-Nano-30B-A3B-Q8
Runtimes: BaseRT
Versions: BaseRT 0.2.6
Quantisations: Q4, Q8
Backends: Metal
Cooldown: off
Conditioning: warmup_only
Decode workload: TG128 @ 1 ctx
Harness schema: basert-benchmark-harness/1

What differs besides the chip

Facet	Apple M5 Max	Apple M1 Max
Model	NVIDIA-Nemotron-3-Nano-30B-A3B	NVIDIA-Nemotron-3-Nano-30B-A3B-Q4, NVIDIA-Nemotron-3-Nano-30B-A3B-Q8
Runtime version	BaseRT 0.2.4	BaseRT 0.2.6
Quantisation	Q4	Q4, Q8
Cooldown	off, on	off
Conditioning	idle_reset_then_warm, warmup_only	warmup_only
Decode workload	TG128	TG128 @ 1 ctx

Like-for-like pairs

Each row is a configuration present on both sides. Values are per-cell medians; ratios read Apple M5 Max relative to Apple M1 Max.

Configuration	Decode A	Decode B	Ratio	Prefill A	Prefill B	Ratio	Runs A / B
NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 BaseRTQ4	179.2	110.3	1.62×	4,942	870	5.68×	6 / 1

Side A · Apple M5 Max

Top 6 of 6 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B BaseRT 0.2.4Q4	Apple M5 Max Metal	184.4 TG128	4,984 PP512	lukas	11 Sep 2026	View Benchmark
basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B BaseRT 0.2.4Q4	Apple M5 Max Metal	183.7 TG128	4,941 PP512	lukas	11 Sep 2026	View Benchmark
basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B BaseRT 0.2.4Q4	Apple M5 Max Metal	180.8 TG128	4,951 PP512	lukas	11 Sep 2026	View Benchmark
basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B BaseRT 0.2.4Q4	Apple M5 Max Metal	177.6 TG128	4,527 PP512	basecompute	10 Sep 2026	View Benchmark
basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B BaseRT 0.2.4Q4	Apple M5 Max Metal	177.0 TG128	4,821 PP512	lukas	11 Sep 2026	View Benchmark
basecompute/NVIDIA-Nemotron-3-Nano-30B-A3B BaseRT 0.2.4Q4	Apple M5 Max Metal	109.2 TG128	4,943 PP512	lukas	11 Sep 2026	View Benchmark

Side B · Apple M1 Max

Top 3 of 3 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
NVIDIA-Nemotron-3-Nano-30B-A3B-Q4 BaseRT 0.2.6Q4	Apple M1 Max Metal	110.3 TG128 @ 1 ctx	870 PP512	basecompute	20 Sep 2026	View Benchmark
NVIDIA-Nemotron-3-Nano-30B-A3B-Q8 BaseRT 0.2.6Q8	Apple M1 Max Metal	61.1 TG128 @ 1 ctx	857 PP512	basecompute	19 Sep 2026	View Benchmark
NVIDIA-Nemotron-3-Nano-30B-A3B-Q8 BaseRT 0.2.6Q8	Apple M1 Max Metal	60.6 TG128 @ 1 ctx	858 PP512	basecompute	19 Sep 2026	View Benchmark