Molmo2 8B — Hardware Requirements & GPU Compatibility
Vision
Molmo2 8B is an 8.7-billion-parameter vision-language model from the Allen Institute for AI (Ai2), built on a Qwen3-8B backbone paired with a SigLIP2 vision encoder. It is tuned for image, video, and multi-image understanding, with particular strength in pointing, counting, and tracking objects across frames rather than just describing them. Its size suits local deployment on a mainstream-to-high-end consumer GPU once quantized, a step down from the 72B Molmo model it succeeds. The model supports a 36,864 token context window, adequate for multi-image prompts and moderate video rather than long documents. It is released under the Apache 2.0 license, allowing unrestricted commercial and research use. Published in December 2025, Molmo2 8B is notable for pixel-level grounding and for tracking objects over time in video.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 16K)Best on 24 GB (Q4_K_M · ~113 tok/s · 36K)Macs from 16 GB (Q4_K_M · ~24 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Molmo2 8B
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/mu2solutions/Molmo2-8B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 16.1 GB free · 16K context
Download: Molmo2-8B-text-Q4_K_M.gguf5.03 GBmu2solutions
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 3080 Ti (12 GB) (4 GB headroom) · Q4_K_M
- Generation speed
- ~102 tok/s
- generation speed
- Cost per 1M output tokens
- $0.33
- per 1M output tokens
How Much VRAM Does Molmo2 8B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 4.3 GB | 8.8 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 36K context · ~153 tok/s | |
| 4.8 GB | 9.4 GB | 3-bit medium — 3-bit medium quantization | Fits: 36K context · ~136 tok/s | |
| 5.8 GB | 10.3 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 36K context · ~113 tok/s | |
| 6.8 GB | 11.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 36K context · ~97 tok/s | |
| 7.8 GB | 12.3 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 36K context · ~85 tok/s | |
| 9.3 GB | 13.8 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 36K context · ~71 tok/s | |
| 17.9 GB | 22.5 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 36K context · ~37 tok/s |
Selected Q4_K_M
4-bit medium quantization7.9GBat 16K context
- Weights
- 5.2 GB
- KV cache
- 2.4 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 16.1 GB free · ~113 tok/s
≥50 tok/s needs ≥446 GB/s · RTX 5090 ~201, RTX 3090 Ti ~113
Which Hardware Can Run Molmo2 8B?
Q4_K_M · 5.8 GBMolmo2 8B (Q4_K_M) requires 5.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 8+ GB is recommended. Using the full 37K context window can add up to 5.1 GB, bringing total usage to 10.9 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti. 151 devices with unified memory can run Molmo2 8B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA Jetson Orin Nano 8GB (Super).
≤ 11 GB
≤ 11 GB tier →fits · 0.1 GB free · up to 16K ctx
16 GPUs · 1 device family
RTX 3080~85Simulate output on RTX 3080RTX 2080 Ti~69Simulate output on RTX 2080 TiRTX 3070 Ti~68Simulate output on RTX 3070 TiOrin Nano~11Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)20 GB
fits · 12.1 GB free · up to 36K ctx
1 GPU
RX 7900 XT~83Simulate output on RX 7900 XT32–40 GB
fits · 24.1 GB free · up to 36K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~201Simulate output on RTX 5090M5 Max36 GB~56Simulate output on Mac Studio M5 Max (36 GB)M532 GB~19Simulate output on MacBook Air 13" M5 (32 GB)Orin~23Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~15Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
How good is Molmo2 8B?
llmrun Score blends public benchmark boards into one number; 100 = the average open model.
Not enough data yet for an llmrun Score.
A Score needs results on enough independent boards. How the Composite Score is built
Benchmarks
This is the single board showing results for this model; badges show who ran the test.
- LMArena Text
LMArena · Oct 2026Arena
Arena1321.4
#95/188 open
Alternatives to Molmo2 8B
OLMo family · 20 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite Switch 4.1 8BIBM | — | 8.4 GB | ~111 | same size |
| ALoDLM 8BAmazon | — | 7.6 GB | ~119 | same size |
| Llm Jp 4.1 8B Thinkingllm-jp | — | 7.6 GB | ~115 | same size |
| Apertus V1.5 8Bswiss-ai | — | 8.5 GB | ~111 | same size |
| LFM2.5 8B A1BLiquid AI | — | 5.6 GB | ~335 | same size |
| Chandradatalab-to | — | 8 GB | ~112 | same size |
Specifications
- Publisher
- Allen AI
- Family
- OLMo
- Parameters
- 8.7B
- Architecture
- Molmo2ForConditionalGeneration
- Context Length
- 36,864 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2025-12-14
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Molmo2 8B need?
Molmo2 8B requires 5.8 GB of VRAM at Q4_K_M, or 17.9 GB at BF16. Full 37K context adds up to 5.1 GB (10.9 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 8.7B × 4.8 bits ÷ 8 = 5.2 GB
KV Cache + Overhead ≈ 0.6 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 ≈ 5.7 GB (at full 37K context)
VRAM usage by quantization
What's the best quantization for Molmo2 8B?
For Molmo2 8B, Q4_K_M (5.8 GB) offers the best balance of quality and VRAM usage. Q5_K_M (6.8 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 4.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Molmo2 8B on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run Molmo2 8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 5.8 GB of usable unified memory (RAM minus macOS overhead).
Can I run Molmo2 8B locally?
Yes — Molmo2 8B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Molmo2 8B?
At Q4_K_M, Molmo2 8B can reach ~828 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~113 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 ÷ 5.8 × 0.65 = ~897 tok/s
Estimated speed at Q4_K_M (5.8 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Molmo2 8B?
At Q4_K_M, the download is about 5.20 GB. The full-precision BF16 version is 17.32 GB. The smallest option (Q2_K) is 3.68 GB.
Which GPUs can run Molmo2 8B?
52 consumer GPUs can run Molmo2 8B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 40 GPUs have plenty of headroom for comfortable inference.
Which devices can run Molmo2 8B?
151 devices with unified memory can run Molmo2 8B at Q4_K_M (5.8 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.