MNN Pocket 0.3B — Hardware Requirements & GPU Compatibility
ChatFunctions
MNN Pocket 0.3B is a 362M-parameter open language model from taobao-mnn. It supports a context window of up to 40,960 tokens. At Q4_K_M it needs about 0.55 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 40K)Best on 24 GB (Q4_K_M · ~1191 tok/s · 40K)Macs from 8 GB (Q4_K_M · ~130 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run MNN Pocket 0.3B
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/taobao-mnn/MNN-Pocket-0.3B-GGUFQ4_K_M fits fully on the RTX 4090 · 23.2 GB free · 16K context
Download: MNN-Pocket-0.3B-Q4_K_M.gguf0.22 GBtaobao-mnn
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (11 GB headroom) · Q4_K_M
- Generation speed
- ~426 tok/s
- generation speed
- Cost per 1M output tokens
- $0.038
- per 1M output tokens
How Much VRAM Does MNN Pocket 0.3B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.5 GB | 0.9 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 40K context · ~1365 tok/s | |
| 0.5 GB | 1 GB | 3-bit medium — 3-bit medium quantization | Fits: 40K context · ~1285 tok/s | |
| 0.5 GB | 1 GB | 4-bit · legacy+ — 4-bit legacy quantization with offset | Fits: 40K context · ~1236 tok/s | |
| 0.6 GB | 1 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 40K context · ~1191 tok/s | |
| 0.6 GB | 1 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 40K context · ~1111 tok/s | |
| 0.6 GB | 1.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 40K context · ~1040 tok/s | |
| 0.7 GB | 1.1 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 40K context · ~950 tok/s | |
| 1.1 GB | 1.5 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 40K context · ~624 tok/s |
Selected Q4_K_M
4-bit medium quantization0.8GBat 16K context
- Weights
- 0.2 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 23.2 GB free · ~1191 tok/s
≥50 tok/s needs ≥42 GB/s · RTX 5090 ~2118, RTX 3090 Ti ~1191
Which Hardware Can Run MNN Pocket 0.3B?
Q4_K_M · 0.6 GBMNN Pocket 0.3B (Q4_K_M) requires 0.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. Using the full 41K context window can add up to 0.6 GB, bringing total usage to 1.1 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run MNN Pocket 0.3B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 7.2 GB free · up to 40K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~899Simulate output on RTX 3080RTX 2080 Ti~728Simulate output on RTX 2080 TiRTX 3070 Ti~719Simulate output on RTX 3070 TiM3*8 GB~130Simulate output on iMac 24" M3 (8 GB)Orin Nano~121Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~87Simulate output on iPhone 1720 GB
fits · 19.2 GB free · up to 40K ctx
1 GPU
RX 7900 XT~873Simulate output on RX 7900 XT32–40 GB
fits · 31.2 GB free · up to 40K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~2118Simulate output on RTX 5090M5 Max36 GB~586Simulate output on Mac Studio M5 Max (36 GB)M532 GB~196Simulate output on MacBook Air 13" M5 (32 GB)Orin~242Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~160Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
5| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite 4.0 350MIBM | — | 1 GB | ~1150 | same size |
| SmolLM 360MHugging Face | — | 0.6 GB | ~1092 | same size |
| LFM2.5 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
| ERNIE 4.5 0.3B PTBaidu | — | 0.7 GB | ~1213 | same size |
| LFM2 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
Specifications
- Publisher
- taobao-mnn
- Parameters
- 362M
- Architecture
- Qwen3ForCausalLM
- Context Length
- 40,960 tokens
- Vocabulary Size
- 65,536
- Release Date
- 2026-09-23
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does MNN Pocket 0.3B need?
MNN Pocket 0.3B requires 0.6 GB of VRAM at Q4_K_M, or 1.1 GB at BF16. Full 41K context adds up to 0.6 GB (1.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 362M × 4.8 bits ÷ 8 = 0.2 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 0.9 GB (at full 41K context)
VRAM usage by quantization
What's the best quantization for MNN Pocket 0.3B?
For MNN Pocket 0.3B, Q4_K_M (0.6 GB) offers the best balance of quality and VRAM usage. Q5_K_M (0.6 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.5 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run MNN Pocket 0.3B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run MNN Pocket 0.3B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.6 GB of usable unified memory (RAM minus macOS overhead).
Can I run MNN Pocket 0.3B locally?
Yes — MNN Pocket 0.3B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is MNN Pocket 0.3B?
At Q4_K_M, MNN Pocket 0.3B can reach ~8727 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~1191 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 0.55 × 0.65 = ~9455 tok/s
Estimated speed at Q4_K_M (0.6 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of MNN Pocket 0.3B?
At Q4_K_M, the download is about 0.22 GB. The full-precision BF16 version is 0.72 GB. The smallest option (Q2_K) is 0.15 GB.
Which GPUs can run MNN Pocket 0.3B?
52 consumer GPUs can run MNN Pocket 0.3B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run MNN Pocket 0.3B?
163 devices with unified memory can run MNN Pocket 0.3B at Q4_K_M (0.6 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.