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: 35
Like-for-like pairs: 2
Runtimes available: 2

Compare

Side A

Side B

Model

Chip

Quantisation A

Quantisation B

Backend

Prefill size

Decode size

Cooldown

23 runs vs 12 runs · 2 like-for-like pairs

Held fixedModel: Qwen3 0.6BPrefill: PP512Decode: TG128

The sides differ in conditioning, decode workload, harness schema. Ratios may reflect benchmark protocol rather than the runtime.

Runtime comparison

BaseRT vs llama.cpp

Ratios read side A relative to side B. Geometric mean over 2 like-for-like pairs.

Decode: 1.41×+41.0%
Prefill: 1.37×+37.4%

Side A

BaseRT

23 runs

5 contributors

Latest 19 Sep 2026

Decode · TG128: 440.2tok/s median
best 708.3 · worst 74.3
Prefill · PP512: 19,595tok/s median
best 52,750 · worst 1,851

Models: Qwen3-0.6B, Qwen3-0.6B-cuda-q4, Qwen3-0.6B-cuda-q4mix, Qwen3-0.6B-cuda-q8, Qwen3-0.6B-Q4, Qwen3-0.6B-Q8
Chips: Apple M1 Max, Apple M1 Pro, Apple M3 Ultra, Apple M4 Max, Apple M4 Pro, Apple M5, Apple M5 Max, Apple M5 Pro, NVIDIA GB10
Versions: BaseRT 0.2.4, BaseRT 0.2.5, BaseRT 0.2.6
Quantisations: Q4, Q8
Backends: CUDA, Metal
Cooldown: off
Conditioning: warmup_only
Decode workload: TG128, TG128 @ 1 ctx
Harness schema: basert-benchmark-harness/1

Side B

llama.cpp

12 runs

2 contributors

Latest 11 Sep 2026

Decode · TG128: 368.2tok/s median
best 508.8 · worst 283.0
Prefill · PP512: 24,169tok/s median
best 26,346 · worst 14,344

Models: Models Qwen Qwen3 0.6B, Qwen3 0.6B Instruct, Qwen3-0.6B
Chips: AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XT (RADV NAVI31), Apple M5 Max, Apple M5 Pro
Versions: llama.cpp b10360 (48d22e295), llama.cpp b10809 (5266f24da)
Quantisations: Q4_K_M, Q8_0
Backends: BLAS + Metal, ROCm, Vulkan
Cooldown: off
Conditioning: runtime_native_warmup
Decode workload: TG128
Harness schema: computearena-measurements/1

What differs besides the runtime

Facet	BaseRT	llama.cpp
Model	Qwen3-0.6B, Qwen3-0.6B-cuda-q4, Qwen3-0.6B-cuda-q4mix, Qwen3-0.6B-cuda-q8, Qwen3-0.6B-Q4, Qwen3-0.6B-Q8	Models Qwen Qwen3 0.6B, Qwen3 0.6B Instruct, Qwen3-0.6B
Chip	Apple M1 Max, Apple M1 Pro, Apple M3 Ultra, Apple M4 Max, Apple M4 Pro, Apple M5, Apple M5 Max, Apple M5 Pro, NVIDIA GB10	AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XT (RADV NAVI31), Apple M5 Max, Apple M5 Pro
Backend	CUDA, Metal	BLAS + Metal, ROCm, Vulkan
Conditioning	warmup_only	runtime_native_warmup
Decode workload	TG128, TG128 @ 1 ctx	TG128
Harness schema	basert-benchmark-harness/1	computearena-measurements/1

Like-for-like pairs

Each row is a configuration present on both sides. Values are per-cell medians; ratios read BaseRT relative to llama.cpp.

Configuration	Decode A	Decode B	Ratio	Prefill A	Prefill B	Ratio	Runs A / B
Qwen3-0.6B Apple M5 Pro	415.8	357.2	1.16×	20,451	14,530	1.41×	6 / 4
Qwen3-0.6B Apple M5 Max	698.5	409.0	1.71×	33,088	24,674	1.34×	3 / 2

Side A · BaseRT

Top 23 of 23 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
basecompute/Qwen3-0.6B BaseRT 0.2.4Q4	Apple M5 Max Metal	708.3 TG128	34,136 PP512	basecompute	10 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.5Q4	Apple M5 Max Metal	698.5 TG128	32,977 PP512	lukas	18 Sep 2026	View Benchmark
Qwen3-0.6B-Q4 BaseRT 0.2.6Q4	Apple M4 Max Metal	583.3 TG128 @ 1 ctx	9,998 PP512	basecompute	19 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.4Q8	Apple M5 Max Metal	549.5 TG128	33,088 PP512	basecompute	10 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.4Q4	Apple M5 Pro Metal	516.8 TG128	20,778 PP512	arki05	11 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.4Q4	Apple M5 Pro Metal	512.0 TG128	20,603 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B-Q4 BaseRT 0.2.6Q4	Apple M3 Ultra Metal	501.7 TG128 @ 1 ctx	12,830 PP512	basecompute	19 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.5Q4	Apple M5 Pro Metal	481.5 TG128 @ 1 ctx	19,595 PP512	isu	18 Sep 2026	View Benchmark
Qwen3-0.6B-Q4 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	458.8 TG128 @ 1 ctx	20,970 PP512	basecompute	19 Sep 2026	View Benchmark
Qwen3-0.6B-cuda-q4 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	456.1 TG128 @ 1 ctx	52,750 PP512	basecompute	19 Sep 2026	View Benchmark
Qwen3-0.6B-Q8 BaseRT 0.2.6Q8	Apple M4 Max Metal	455.8 TG128 @ 1 ctx	9,749 PP512	basecompute	19 Sep 2026	View Benchmark
Qwen3-0.6B-Q8 BaseRT 0.2.6Q8	Apple M3 Ultra Metal	440.2 TG128 @ 1 ctx	12,494 PP512	basecompute	19 Sep 2026	View Benchmark
Qwen3-0.6B-cuda-q4mix BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	413.2 TG128 @ 1 ctx	49,865 PP512	basecompute	19 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.4Q8	Apple M5 Pro Metal	350.2 TG128	20,537 PP512	arki05	11 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.4Q8	Apple M5 Pro Metal	349.8 TG128	20,366 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B-cuda-q8 BaseRT 0.2.6Q4	NVIDIA GB10 CUDA	315.8 TG128 @ 1 ctx	51,373 PP512	basecompute	19 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.4Q8	Apple M5 Pro Metal	311.0 TG128	18,474 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B-Q8 BaseRT 0.2.6Q8	Apple M4 Pro Metal	284.7 TG128 @ 1 ctx	4,787 PP512	basecompute	19 Sep 2026	View Benchmark
basecompute/Qwen3-0.6B BaseRT 0.2.6Q4	Apple M1 Pro Metal	268.8 TG128 @ 1 ctx	2,847 PP512	isu	19 Sep 2026	View Benchmark
Qwen3-0.6B-Q4 BaseRT 0.2.6Q4	Apple M4 Pro Metal	208.6 TG128 @ 1 ctx	3,571 PP512	basecompute	19 Sep 2026	View Benchmark
Qwen/Qwen3-0.6B BaseRT 0.2.4Q4	Apple M5 Metal	172.7 TG128	4,997 PP512	skogul97	11 Sep 2026	View Benchmark
Qwen3-0.6B-Q4 BaseRT 0.2.6Q4	Apple M1 Max Metal	100.8 TG128 @ 1 ctx	1,851 PP512	basecompute	19 Sep 2026	View Benchmark
Qwen3-0.6B-Q8 BaseRT 0.2.6Q8	Apple M1 Max Metal	74.3 TG128 @ 1 ctx	2,225 PP512	basecompute	19 Sep 2026	View Benchmark

Side B · llama.cpp

Top 12 of 12 runs by decode throughput.

Model / format	Chip / backend	Decode	Prefill	Contributor	Date	Report
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	AMD Radeon RX 7900 XT (RADV NAVI31) Vulkan	508.8 TG128	26,137 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	AMD Radeon RX 7900 XT (RADV NAVI31) Vulkan	491.9 TG128	25,964 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q8_0	AMD Radeon RX 7900 XT (RADV NAVI31) Vulkan	479.4 TG128	26,346 PP512	arki05	11 Sep 2026	View Benchmark
Models Qwen Qwen3 0.6B llama.cpp b10360 (48d22e295)Q4_K_M	Apple M5 Max BLAS + Metal	439.1 TG128	24,365 PP512	basecompute	10 Sep 2026	View Benchmark
Qwen3 0.6B Instruct llama.cpp b10360 (48d22e295)Q8_0	Apple M5 Max BLAS + Metal	379.0 TG128	24,983 PP512	basecompute	10 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	AMD Radeon RX 7900 XT ROCm	368.6 TG128	23,972 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	AMD Radeon RX 7900 XT ROCm	367.7 TG128	23,718 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	Apple M5 Pro BLAS + Metal	358.3 TG128	14,509 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	Apple M5 Pro BLAS + Metal	358.1 TG128	14,344 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q4_K_M	Apple M5 Pro BLAS + Metal	356.2 TG128	14,552 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q8_0	AMD Radeon RX 7900 XT ROCm	313.7 TG128	24,691 PP512	arki05	11 Sep 2026	View Benchmark
Qwen3-0.6B llama.cpp b10809 (5266f24da)Q8_0	Apple M5 Pro BLAS + Metal	283.0 TG128	14,942 PP512	arki05	11 Sep 2026	View Benchmark