Downloads · 30 days
147
28% of all-time downloads
gemma-4-26B-A4B-it-oQ4e-mtp
mlx-works/gemma-4-26B-A4B-it-oQ4e-mtp
gemma-4-26B-A4B-it-oQ4e-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 v0.5.7) mixed-precision quantization.
All-time downloads
527
Public
Parameters
26.2B
16.7 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors16.6 GB · 100%
Parameter types
How the weights are stored.
U3225.2B · 96%
Model card
From the Hugging Face model README
gemma-4-26B-A4B-it-oQ4e-mtp
This model was quantized using oQ (oMLX v0.5.7) mixed-precision quantization.
Base model: google/gemma-4-26B-A4B-it
MTP head: google/gemma-4-26B-A4B-it-assistant (merged into single checkpoint)
Quantization details
- Model type: gemma4
- Bits: 4
- Group size: 64
- Format: MLX safetensors
- MTP: Merged from assistant checkpoint (Lightning MTP)
- Calibration: oQ4e (enhanced, imatrix-based; oqe_code_multilingual)
Environment
- Hardware: M5 MacBook Air 32GB
- Inference Framework: oMLX v0.5.7
- 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 | 1070.5 | 25.97 | 956.6 tok/s | 38.8 tok/s | 4.380 | 263.0 tok/s | 14.40 GB |
| pp4096/tg128 | 4660.0 | 27.20 | 879.0 tok/s | 37.1 tok/s | 8.138 | 519.0 tok/s | 14.71 GB |
Continuous Batching (pp1024 / tg128)
| Batch | tg TPS | Speedup | pp TPS | pp TPS/req | TTFT(ms) | E2E(s) |
|---|---|---|---|---|---|---|
| 1x | 38.8 tok/s | 1.00x | 956.6 tok/s | 956.6 tok/s | 1070.5 | 4.380 |
| 2x | 54.2 tok/s | 1.40x | 796.2 tok/s | 398.1 tok/s | 2572.2 | 7.294 |
| 4x | 70.5 tok/s | 1.82x | 809.0 tok/s | 202.3 tok/s | 4825.8 | 12.330 |
Intelligence Benchmark
Note: Each benchmark round tests only 30 questions. Results are for reference only.
| Benchmark | Accuracy | Correct | Total | Time(s) | Think |
|---|---|---|---|---|---|
| MMLU | 70.0% | 21 | 30 | 30.2 | No |
| TRUTHFULQA | 86.7% | 26 | 30 | 15.1 | No |
| GSM8K | 93.3% | 28 | 30 | 150.3 | No |
| MATHQA | 16.7% | 5 | 30 | 102.9 | No |
| HUMANEVAL | 96.7% | 29 | 30 | 275.1 | No |
- Library
- mlx
- Type
- gemma4
- Created
- Aug 15, 2026
- Updated
- Aug 16, 2026