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: 7

Compare

Side A

Side B

Model

Runtime

Runtime version

Quantisation

Backend

Prefill size

Decode size

Cooldown

5 runs vs 4 runs · 1 like-for-like pair

Held fixedModel: Llama 3.2 1B InstructPrefill: PP512Decode: TG128

Chip comparison

NVIDIA GB10 vs Apple M1 Max

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

Decode: 0.74×-26.3%
Prefill: 10.72×+972%

Side A

NVIDIA GB10

5 runs

1 contributor

Latest 9 Oct 2026

Decode · TG128: 173.7tok/s median
best 261.3 · worst 169.5
Prefill · PP512: 35,941tok/s median
best 36,737 · worst 33,397

Models: Llama-3.2-1B-Instruct-cuda-q4mix, Llama-3.2-1B-Instruct-cuda-q8, Llama-3.2-1B-Instruct-Q4, Llama-3.2-1B-Instruct-Q8
Runtimes: BaseRT
Versions: BaseRT 0.2.6, BaseRT 0.3.0
Quantisations: Q4
Backends: CUDA
Cooldown: off
Conditioning: warmup_only
Decode workload: TG128 @ 1 ctx
Harness schema: basert-benchmark-harness/1

Side B

Apple M1 Max

4 runs

1 contributor

Latest 9 Oct 2026

Decode · TG128: 235.5tok/s median
best 242.1 · worst 229.4
Prefill · PP512: 3,353tok/s median
best 3,372 · worst 3,333

Models: Llama-3.2-1B-Instruct-Q4, Llama-3.2-1B-Instruct-Q8
Runtimes: BaseRT
Versions: BaseRT 0.2.6, BaseRT 0.3.0
Quantisations: Q4
Backends: Metal
Cooldown: off
Conditioning: warmup_only
Decode workload: TG128 @ 1 ctx
Harness schema: basert-benchmark-harness/1

What differs besides the chip

Facet	NVIDIA GB10	Apple M1 Max
Model	Llama-3.2-1B-Instruct-cuda-q4mix, Llama-3.2-1B-Instruct-cuda-q8, Llama-3.2-1B-Instruct-Q4, Llama-3.2-1B-Instruct-Q8	Llama-3.2-1B-Instruct-Q4, Llama-3.2-1B-Instruct-Q8
Backend	CUDA	Metal

Like-for-like pairs

Each row is a configuration present on both sides. Values are per-cell medians; ratios read NVIDIA GB10 relative to Apple M1 Max.

Configuration	Decode A	Decode B	Ratio	Prefill A	Prefill B	Ratio	Runs A / B
Llama-3.2-1B-Instruct BaseRTQ4	173.7	235.5	0.74×	35,941	3,353	10.72×	5 / 4

Side A · NVIDIA GB10

Top 5 of 5 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
Llama-3.2-1B-Instruct-cuda-q4mix BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	261.3 TG128 @ 1 ctx	36,647 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-Q4 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	195.3 TG128 @ 1 ctx	33,397 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-cuda-q8 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	173.7 TG128 @ 1 ctx	34,296 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-Q8 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	172.8 TG128 @ 1 ctx	35,941 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-cuda-q8 BaseRT 0.3.0Q4	NVIDIA GB10 CUDA	169.5 TG128 @ 1 ctx	36,737 PP512	basecompute	9 Oct 2026	View Benchmark

Side B · Apple M1 Max

Top 4 of 4 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
Llama-3.2-1B-Instruct-Q4 BaseRT 0.2.6Q4	Apple M1 Max Metal	242.1 TG128 @ 1 ctx	3,333 PP512	basecompute	20 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-Q4 BaseRT 0.3.0Q4	Apple M1 Max Metal	238.3 TG128 @ 1 ctx	3,348 PP512	basecompute	9 Oct 2026	View Benchmark
Llama-3.2-1B-Instruct-Q8 BaseRT 0.3.0Q4	Apple M1 Max Metal	232.7 TG128 @ 1 ctx	3,357 PP512	basecompute	9 Oct 2026	View Benchmark
Llama-3.2-1B-Instruct-Q8 BaseRT 0.2.6Q4	Apple M1 Max Metal	229.4 TG128 @ 1 ctx	3,372 PP512	basecompute	20 Sep 2026	View Benchmark

Compare

Loading community benchmarks…

Facet

NVIDIA GB10

Apple M1 Max

Model

Llama-3.2-1B-Instruct-cuda-q4mix, Llama-3.2-1B-Instruct-cuda-q8, Llama-3.2-1B-Instruct-Q4, Llama-3.2-1B-Instruct-Q8

Llama-3.2-1B-Instruct-Q4, Llama-3.2-1B-Instruct-Q8

Backend

CUDA

Metal

Configuration

Decode A

Decode B

Ratio

Prefill A

Prefill B

Ratio

Runs A / B

Llama-3.2-1B-Instruct

BaseRTQ4

173.7

235.5

0.74×

35,941

3,353

10.72×

5 / 4

Side A · NVIDIA GB10

Top 5 of 5 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
Llama-3.2-1B-Instruct-cuda-q4mix BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	261.3 TG128 @ 1 ctx	36,647 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-Q4 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	195.3 TG128 @ 1 ctx	33,397 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-cuda-q8 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	173.7 TG128 @ 1 ctx	34,296 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-Q8 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	172.8 TG128 @ 1 ctx	35,941 PP512	basecompute	19 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-cuda-q8 BaseRT 0.3.0Q4	NVIDIA GB10 CUDA	169.5 TG128 @ 1 ctx	36,737 PP512	basecompute	9 Oct 2026	View Benchmark

Side B · Apple M1 Max

Top 4 of 4 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
Llama-3.2-1B-Instruct-Q4 BaseRT 0.2.6Q4	Apple M1 Max Metal	242.1 TG128 @ 1 ctx	3,333 PP512	basecompute	20 Sep 2026	View Benchmark
Llama-3.2-1B-Instruct-Q4 BaseRT 0.3.0Q4	Apple M1 Max Metal	238.3 TG128 @ 1 ctx	3,348 PP512	basecompute	9 Oct 2026	View Benchmark
Llama-3.2-1B-Instruct-Q8 BaseRT 0.3.0Q4	Apple M1 Max Metal	232.7 TG128 @ 1 ctx	3,357 PP512	basecompute	9 Oct 2026	View Benchmark
Llama-3.2-1B-Instruct-Q8 BaseRT 0.2.6Q4	Apple M1 Max Metal	229.4 TG128 @ 1 ctx	3,372 PP512	basecompute	20 Sep 2026	View Benchmark