Gemma 3 4B (Unsloth)
Gemma 3 4B is the most unusual Nano-class variant in the CrucibleMark portfolio: a multimodal Google DeepMind model with text and image processing at just 4 billion parameters. 128,000 tokens of context, Gemma terms of use, locally deployable as Unsloth-GGUF — the most compact multimodal member of the Gemma 3 family.
- Restricted Weights
- Nano
- llama.cpp
- Text
- Vision
- Instruction-Tuned
- Restricted-Weights
- Real-Time
Sovereign Risk: LOW TODO
Key metrics
Score · Latency · Cost · Quality
- Total Score Bronze
- 61.52
- Routine
- 37.9
- Reasoning
- 23.63
- LLM Judge Avg
- 2.79 / 5
- 100 Coverage
- Avg Task Duration
- 18.29s
- Real-Time
- Token Rate
- 44.46tok/s
- Output Rate
- P95 Latency
- 31.66s
- Top 5 %
- Total Tokens
- 58500
- Output Volume
- Cost per 1K
- $0
- USD / 1K Requests
- Benchmark Cost
- $0
- Total · 58500 tok
Benchmark modules
10 modules · weighted · vs. model median & top performer
Gemma 3 4B (Unsloth)
Best model
Ø All models
Code Quality
58.8
CLI Benchmark
71.12
Logical Reasoning
58.73
UX Writing
60.15
Documentation
47.15
Content Transform.
68.52
Cultural Intelligence
71
Synthesis Quality
Tool Execution
ToolUse Score
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.