Downloads · 30 days
28
13% of all-time downloads
gemma-4-E2B-it-oQ6-mtp
mlx-works/gemma-4-E2B-it-oQ6-mtp
gemma-4-E2B-it-oQ6-mtp is a machine learning model from mlx-works. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx.
This model was quantized using oQ (oMLX v0.5.5) mixed-precision quantization.
All-time downloads
208
Public
Parameters
5.1B
5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors5 GB · 99%
Parameter types
How the weights are stored.
U324.7B · 91%
Model card
From the Hugging Face model README
gemma-4-E2B-it-oQ6-mtp
This model was quantized using oQ (oMLX v0.5.5) mixed-precision quantization.
Base model: google/gemma-4-E2B-it
MTP head: google/gemma-4-E2B-it-assistant (merged into single checkpoint)
Quantization details
- Model type: gemma4
- Bits: 6
- Group size: 64
- Format: MLX safetensors
- MTP: Merged from assistant checkpoint (Lightning MTP)
Environment
- Hardware: M5 MacBook Air 32GB
- Inference Framework: oMLX v0.5.5
- Max Concurrent Requests: 4
- Settings:
- Thinking: Disabled
- TurboQuant KV Cache: Enabled (4-bit)
- Lightning MTP: Enabled (key speed improvement)
Performance Benchmarks
Note: Results are for reference only and may vary depending on hardware, software configuration, and workload.
Single Request Results
| Test | TTFT(ms) | TPOT(ms) | pp TPS | tg TPS | E2E(s) | Throughput | Peak Mem |
|---|---|---|---|---|---|---|---|
| pp1024/tg128 | 186.4 | 18.47 | 5494.5 tok/s | 54.6 tok/s | 2.547 | 452.4 tok/s | 4.16 GB |
| pp4096/tg128 | 1158.3 | 19.73 | 3536.3 tok/s | 51.1 tok/s | 3.691 | 1144.5 tok/s | 4.19 GB |
Continuous Batching (pp1024 / tg128)
| Batch | tg TPS | Speedup | pp TPS | pp TPS/req | TTFT(ms) | E2E(s) |
|---|---|---|---|---|---|---|
| 1x | 54.6 tok/s | 1.00x | 5494.5 tok/s | 5494.5 tok/s | 186.4 | 2.547 |
| 2x | 100.4 tok/s | 1.84x | 3047.9 tok/s | 1524.0 tok/s | 671.8 | 3.221 |
| 4x | 176.5 tok/s | 3.23x | 3686.0 tok/s | 921.5 tok/s | 962.7 | 4.013 |
Intelligence Benchmark
Note: Each benchmark round tests only 30 questions. Results are for reference only.
| Benchmark | Accuracy | Correct | Total | Time(s) | Think |
|---|---|---|---|---|---|
| MMLU | 56.7% | 17 | 30 | 10.9 | No |
| TRUTHFULQA | 46.7% | 14 | 30 | 6 | No |
| GSM8K | 93.3% | 28 | 30 | 74.8 | No |
| MATHQA | 13.3% | 4 | 30 | 36.5 | No |
| HUMANEVAL | 80.0% | 24 | 30 | 106.3 | No |
- Library
- mlx
- Type
- gemma4
- Created
- Aug 5, 2026
- Updated
- Aug 16, 2026