Granite 4.0 H Tiny — Hardware Requirements & GPU Compatibility
ChatFunctions
Granite 4.0 H Tiny is a 6.9B-parameter open language model from IBM in the Granite family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 4.63 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 · 32K)Best on 24 GB (Q4_K_M · ~271 tok/s · 128K)Macs from 8 GB (IQ3_XS · ~56 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Granite 4.0 H Tiny
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/ibm-granite/granite-4.0-h-tiny-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 18.2 GB free · 16K context
Download: granite-4.0-h-tiny-Q4_K_M.gguf4.23 GBibm-granite
Ways to runOther ways
each opens step-by-stepRun in cloud
Your own private copy · Q4_K_M · 24 GB GPU
- The easiest way to try this model
- Running in a few minutes, easy setup
- Chat right here or connect any app
$5+ bonus for new users
Runs on RTX 3060 12GB (6 GB headroom) · Q4_K_M
- Generation speed
- ~157 tok/s
- generation speed
- Cost per 1M output tokens
- $0.11
- per 1M output tokens
How Much VRAM Does Granite 4.0 H Tiny Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 3.4 GB | 5.9 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~303 tok/s | |
| 3.5 GB | 6 GB | 3-bit small — 3-bit small quantization | Fits: 128K context · ~301 tok/s | |
| 3.9 GB | 6.4 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~291 tok/s | |
| 3.9 GB | 6.5 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 128K context · ~288 tok/s | |
| 4.6 GB | 7.2 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~271 tok/s | |
| 5.4 GB | 7.9 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~254 tok/s | |
| 6.2 GB | 8.7 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~238 tok/s | |
| 7.4 GB | 9.9 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~218 tok/s |
Selected Q4_K_M
4-bit medium quantization5.8GBat 16K context
- Weights
- 4.2 GB
- KV cache
- 1.3 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 18.2 GB free · ~271 tok/s
≥50 tok/s needs ≥84 GB/s · RTX 5090 ~329, RTX 3090 Ti ~271
Which Hardware Can Run Granite 4.0 H Tiny?
Q4_K_M · 4.6 GBGranite 4.0 H Tiny (Q4_K_M) requires 4.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 7+ GB is recommended. Using the full 131K context window can add up to 10.6 GB, bringing total usage to 15.2 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 151 devices with unified memory can run Granite 4.0 H Tiny, including NVIDIA DGX Spark, NVIDIA Jetson AGX Thor Developer Kit.
≤ 11 GB
≤ 11 GB tier →fits · 2.2 GB free · up to 32K ctx
16 GPUs · 2 device families
RTX 3080~239Simulate output on RTX 3080RTX 2080 Ti~215Simulate output on RTX 2080 TiRTX 3070 Ti~214Simulate output on RTX 3070 TiOrin Nano~59Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~35Simulate output on iPhone 1720 GB
fits · 14.2 GB free · up to 128K ctx
1 GPU
RX 7900 XT~100Simulate output on RX 7900 XT32–40 GB
fits · 26.2 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~329Simulate output on RTX 5090M5 Max36 GB~91Simulate output on Mac Studio M5 Max (36 GB)M532 GB~59Simulate output on MacBook Air 13" M5 (32 GB)Orin~105Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~66Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Granite 4.0 H Tiny
Granite family · 28 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Olmo 3 7B InstructAllen AI | — | 8.4 GB | ~114 | same size |
| Mathstral 7B v0.1Mistral AI | — | 6.8 GB | ~133 | same size |
| Mistral 7B Instruct v0.3Mistral AI | — | 6.8 GB | ~133 | same size |
| Phi 3 Small 8k InstructMicrosoft | — | 16.2 GB | ~43 | same size |
| CodeQwen1.5 7BAlibaba | — | 5.7 GB | ~137 | same size |
| Starling LM 7B BetaNexusflow | — | 5.7 GB | ~133 | same size |
Specifications
- Publisher
- IBM
- Family
- Granite
- Parameters
- 6.9B
- Active parameters
- 1.5B per token
- Architecture
- GraniteMoeHybridForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 100,352
- Release Date
- 2025-09-16
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Granite 4.0 H Tiny need?
Granite 4.0 H Tiny requires 4.6 GB of VRAM at Q4_K_M, or 14.3 GB at BF16. Full 131K context adds up to 10.6 GB (15.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 6.9B × 4.8 bits ÷ 8 = 4.2 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 ≈ 11 GB (at full 131K context)
VRAM usage by quantization
What's the best quantization for Granite 4.0 H Tiny?
For Granite 4.0 H Tiny, Q4_K_M (4.6 GB) offers the best balance of quality and VRAM usage. Q4_K_L (4.7 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 2.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Granite 4.0 H Tiny on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Granite 4.0 H Tiny, the smallest at IQ3_XS. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 4.6 GB of usable unified memory (RAM minus macOS overhead), from Mac Mini M2 Pro (16 GB).
Can I run Granite 4.0 H Tiny locally?
Yes — Granite 4.0 H Tiny can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Granite 4.0 H Tiny?
At Q4_K_M, Granite 4.0 H Tiny can reach ~122 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~271 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 → 1.0 GB active ÷ (8000 × 0.65) = 0.188 ms, plus 40 layers × 0.055 ms = 2.200 ms, so 1000 ÷ 2.388 ms = ~419 tok/s
Estimated speed at Q4_K_M (4.6 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Granite 4.0 H Tiny?
At Q4_K_M, the download is about 4.16 GB. The full-precision BF16 version is 13.88 GB. The smallest option (IQ2_XXS) is 1.91 GB.
Which GPUs can run Granite 4.0 H Tiny?
52 consumer GPUs can run Granite 4.0 H Tiny 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 Granite 4.0 H Tiny?
151 devices with unified memory can run Granite 4.0 H Tiny at Q4_K_M (4.6 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.