Tiny Mixtral — Hardware Requirements & GPU Compatibility
Chat
Tiny Mixtral is a 247M-parameter open language model from TitanML in the Mixtral family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 0.45 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 · 128K)Best on 24 GB (Q4_K_M · ~2329 tok/s · 128K)Macs from 8 GB (Q4_K_M · ~301 tok/s)
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Tiny Mixtral
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/tensorblock/tiny-mixtral-GGUFQ4_K_M fits fully on the RTX 4090 · 23.5 GB free · 16K context
Download: tiny-mixtral-Q4_K_M.gguf0.15 GBtensorblock
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (11 GB headroom) · Q4_K_M
- Generation speed
- ~996 tok/s
- generation speed
- Cost per 1M output tokens
- $0.017
- per 1M output tokens
How Much VRAM Does Tiny Mixtral Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.4 GB | 0.5 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~2494 tok/s | |
| 0.4 GB | 0.5 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~2451 tok/s | |
| 0.5 GB | 0.5 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~2329 tok/s | |
| 0.5 GB | 0.5 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~2219 tok/s | |
| 0.5 GB | 0.6 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~2119 tok/s | |
| 0.6 GB | 0.6 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~1999 tok/s | |
| 0.8 GB | 0.9 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 128K context · ~1475 tok/s |
Selected Q4_K_M
4-bit medium quantization0.5GBat 16K context
- Weights
- 0.2 GB
- KV cache
- 0.0 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 23.5 GB free · ~2329 tok/s
≥50 tok/s needs ≥16 GB/s · RTX 5090 ~3452, RTX 3090 Ti ~2329
Which Hardware Can Run Tiny Mixtral?
Q4_K_M · 0.5 GBTiny Mixtral (Q4_K_M) requires 0.5 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 131K context window can add up to 0.3 GB, bringing total usage to 0.7 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Tiny Mixtral, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 7.5 GB free · up to 128K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~1875Simulate output on RTX 3080RTX 2080 Ti~1581Simulate output on RTX 2080 TiRTX 3070 Ti~1564Simulate output on RTX 3070 TiM3*8 GB~301Simulate output on iMac 24" M3 (8 GB)Orin Nano~306Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~209Simulate output on iPhone 1720 GB
fits · 19.5 GB free · up to 128K ctx
1 GPU
RX 7900 XT~1196Simulate output on RX 7900 XT32–40 GB
fits · 31.5 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~3452Simulate output on RTX 5090M5 Max36 GB~953Simulate output on Mac Studio M5 Max (36 GB)M532 GB~426Simulate output on MacBook Air 13" M5 (32 GB)Orin~595Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~387Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Tiny Mixtral
Mixtral family · 7 models →Similar models
3| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| LFM2.5 230MLiquid AI | — | 0.6 GB | ~1310 | same size |
| SmolVLM2 256M Video InstructHugging Face | — | 0.6 GB | ~1310 | same size |
| SmolVLM 256M InstructHugging Face | — | 0.6 GB | ~1310 | same size |
Specifications
- Publisher
- TitanML
- Family
- Mixtral
- Parameters
- 247M
- Active parameters
- 0.1B per token
- Architecture
- MixtralForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 32,000
- Release Date
- 2024-04-24
Frequently Asked Questions
How much VRAM does Tiny Mixtral need?
Tiny Mixtral requires 0.5 GB of VRAM at Q4_K_M, or 0.8 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 247M × 4.8 bits ÷ 8 = 0.1 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 ≈ 0.6 GB (at full 131K context)
VRAM usage by quantization
What's the best quantization for Tiny Mixtral?
For Tiny Mixtral, Q4_K_M (0.5 GB) offers the best balance of quality and VRAM usage. Q5_K_M (0.5 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Tiny Mixtral on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Tiny Mixtral. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.5 GB of usable unified memory (RAM minus macOS overhead).
Can I run Tiny Mixtral locally?
Yes — Tiny Mixtral can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Tiny Mixtral?
At Q4_K_M, Tiny Mixtral can reach ~2254 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~2329 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 0.2 GB active ÷ (8000 × 0.65) = 0.04024 ms, plus 2 layers × 0.055 ms = 0.11000 ms, so 1000 ÷ 0.15024 ms = ~6656 tok/s
Estimated speed at Q4_K_M (0.5 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Tiny Mixtral?
At Q4_K_M, the download is about 0.15 GB. The full-precision BF16 version is 0.49 GB. The smallest option (Q2_K) is 0.10 GB.
Which GPUs can run Tiny Mixtral?
52 consumer GPUs can run Tiny Mixtral 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 Tiny Mixtral?
163 devices with unified memory can run Tiny Mixtral at Q4_K_M (0.5 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.