REAL LOCAL SUITES · CODING V2 + COMPREHENSIVE · SUITES NOT MIXED
HHeavryBenchLocal suites · Real results
FICTIONAL · DEMO ONLY

Fictional Lab / Demo A

Model A Q4

GGUF (simulated) · Q4 (simulated)

HeavryBench Score80.7mock / 100

CAPABILITY PROFILE

Mock subscores

Reasoning
81
Coding
79
Chinese
85
Instruction following
82.7
Structured output
81.7

PERFORMANCE

No measurements

Generation speed
— Not collected
TTFT
— Not collected
Peak RAM
— Not collected
Peak VRAM
— Not collected

PUBLIC SET

Raw answer evidence

demo-public-json-001 · NOT JUDGED

DEMO ONLY: Return a JSON object with keys status and count.

[MOCK ANSWER — NOT MODEL OUTPUT] {"status":"ok","count":1}

REPRODUCIBILITY

Unverified demo record

Run ID
HB-20260907-DEMO02
Suite
HeavryBench Demo v0
Backend
mock 0.1.0
Seed
42
Command
heavrybench run model-a-q4 --backend mock
SHA256
not_collected