Phi-4 Mini (Unsloth)
3.8B dense parameters, synthetic training data, and a pronounced reasoning specialization: Phi-4-mini is Microsoft’s compact model for math, logic, and structured output. MIT license, 128,000 tokens of context, locally deployable as an Unsloth GGUF — not a broad generalist, but a specialist at Nano scale.
- Open Weights
- Nano
- llama.cpp
- Text
- Instruction-Tuned
- Real-Time
Sovereign Risk: LOW Microsoft publishes Phi-4-mini under the MIT License as an Open Weights model, and Unsloth provides the local GGUF packaging for offline inference. The operational provenance risk is low, as the model can be run entirely locally without provider egress.
Key metrics
Score · Latency · Cost · Quality
- Total Score Bronze
- 58.41
- Routine
- 34.53
- Reasoning
- 23.88
- LLM Judge Avg
- 2.79 / 5
- 100 Coverage
- Avg Task Duration
- 17.44s
- Real-Time
- Token Rate
- 46.13tok/s
- Output Rate
- P95 Latency
- 20.51s
- Top 5 %
- Total Tokens
- 62900
- Output Volume
- Cost per 1K
- $0
- USD / 1K Requests
- Benchmark Cost
- $0
- Total · 62900 tok
Benchmark modules
10 modules · weighted · vs. model median & top performer
Phi-4 Mini (Unsloth)
Best model
Ø All models
Code Quality
56.7
CLI Benchmark
78.89
Logical Reasoning
61.09
UX Writing
57.75
Documentation
50.85
Content Transform.
53.71
Cultural Intelligence
52.3
Synthesis Quality
47.5
Tool Execution
88.33
ToolUse Score
66.25
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 / Standard).
- 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.