Benchmark
Submitted byarki05
Public report
This is the public version of the benchmark. It includes model, runtime, protocol, and throughput information, without telemetry, raw samples, or signing metadata.
Compare throughput at the same token counts. A signed report does not independently verify benchmark execution.
Model
- Canonical ID
- Qwen/Qwen3-4B-Instruct-2507
- Identity resolution
- artifact_verified
- Identity verification
- verified
- Verified at
- Name
- Qwen3-4B-Instruct-2507
- Architecture
- qwen3
- Quantization
- Q8_0
- Artifact
- Provider
- huggingface
- Repo ID
- unsloth/Qwen3-4B-Instruct-2507-GGUF
- Revision
- a06e946bb6b655725eafa393f4a9745d460374c9
- Path
- Qwen3-4B-Instruct-2507-Q8_0.gguf
- SHA256
- 391c1e410fd9f4cf2de2b510273b56a84c19ce18f4fa3bfb3774031dac4ef068
Runtime
- Name
- llama-cpp
- Version
- b10809 (5266f24da)
Benchmark
- Mode
- text
- Chip
- Apple M5 Pro
- Backend
- BLAS,MTL
- Protocol
- Report schema
- computearena-benchmark/1
- Benchmark schema
- computearena-measurements/1
- Throughput schema
- llama-bench-independent-pp-tg/1
- Decode initial context tokens
- 0
- Conditioning schema
- computearena-conditioning/1
- Conditioning mode
- runtime_native_warmup
Report metadata
- Schema
- computearena-public-benchmark/1
- Submission ID
- 165fb96e-1c61-46e3-a598-87e900cbb0aa
- Run ID
- 0336711f1bda60c806561e2c4d9702ac
- Benchmark date
- (Unix ms: 1789107301011)
- Submitted by
- arki05