Qwen 3 14B
Qwen 3 14B is Alibaba’s open-weights model for general language tasks and reasoning with an optional thinking mode. The Q6 quantization is designed for efficient local operation without a cloud connection; the context window covers 128,000 tokens. Fully commercially usable under the Apache 2.0 license.
- Open Weights
- Desktop
- llama.cpp
- Text
- Interactive
Sovereign Risk: LOW The model is operated locally without a cloud connection; CLOUD Act and data transfer risks do not apply to purely local inference. The sovereign risk refers to the weights provenance, not to active data transmission.
Key metrics
Score · Latency · Cost · Quality
- Total Score Silver
- 67.66
- Routine
- 41.11
- Reasoning
- 26.55
- LLM Judge Avg
- 3.4 / 5
- 100 Coverage
- Avg Task Duration
- 34.3s
- Interactive
- Token Rate
- 23.87tok/s
- Output Rate
- P95 Latency
- 87.96s
- Top 5 %
- Total Tokens
- 55900
- Output Volume
- Cost per 1K
- $0
- USD / 1K Requests
- Benchmark Cost
- $0
- Total · 55900 tok
Benchmark modules
10 modules · weighted · vs. model median & top performer
Qwen 3 14B
Best model
Ø All models
Code Quality
67.8
CLI Benchmark
91.67
Logical Reasoning
63.09
UX Writing
62.55
Documentation
60.85
Content Transform.
70.38
Cultural Intelligence
71.3
Synthesis Quality
41.67
Tool Execution
90
ToolUse Score
65.67
Benchmark Cost
$0
Token efficiency & latency
Consumption per module vs. model median
Token consumption per module
Performance profile
Key metrics – legend
Score
- Total Score
- Overall score derived from Routine and Reasoning scores. Basis for the fleet rank and tier assignment (Gold / Silver / Bronze).
- Routine
- Score for standard tasks: code generation, text transformation, structured outputs. Weighted by task complexity.
- Reasoning
- Score for analytical and logical tasks: multi-step problem solving, mathematical reasoning, structured argumentation.
- Rank
- Position in the fleet comparison, sorted by Total Score. Refers to the snapshot at the time of testing.
LLM Judge
- LLM Judge Ø
- Average quality rating by an independent judge model on a scale of 1–5.
- Coverage
- Share of tasks for which a judge rating is available. 100% means all responses were evaluated.
Latency & speed
- Avg. Task Duration
- Average time per task from the first to the last token of the response, measured across all benchmark tasks.
- Token Rate
- Output speed in tokens per second. Relevant for interactive applications and real-time streaming.
- P95 Latency
- 95th percentile of response times — shows outliers. Lower values mean fewer long wait times in edge cases.
Cost
- Cost per 1K
- API price per 1,000 output tokens in US dollars. Basis for cost calculation in production use.
- Benchmark Cost
- Total cost for the complete benchmark run of this model.