llmrun
InfinimindCreationsK2HorizonForCausalLM

K2 Horizon MoVA 36B A4B Uncensored — Hardware Requirements & GPU Compatibility

Chat

K2 Horizon MoVA 36B A4B Uncensored is a 37.4B-parameter open language model from InfinimindCreations. It supports a context window of up to 524,288 tokens. At Q4_K_M it needs about 23.02 GB of VRAM — see which GPUs and Macs can run it below.

37.4B params · MoE4B active524K contextApache 2.0149 downloads2 likes
Based on K2 Horizon MoVA 36B A4B
B

Decent on a 24 GB GPU

Runs on24 GB Q4_K_M · ~156 tok/s · 8KMacs from24 GB Q2_K · ~56 tok/s

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run K2 Horizon MoVA 36B A4B Uncensored

One recommended way. Every other method, OS and quant is in the step-by-step guide.

Recommended

vLLM
one command · needs vLLM installed
$vllm serve InfinimindCreations/K2-Horizon-MoVA-36B-A4B-uncensored --max-model-len 16384

vLLM repo with its own quantisation

Download: Weights on Hugging Face

install, OS, other quants

Other ways

each opens step-by-step

Run in cloud

Fits on RTX A6000 (48 GB) (23 GB headroom) · Q4_K_M

Generation speed
~132 tok/s
generation speed
Cost per 1M output tokens
$0.69
per 1M output tokens
Deployon RunPodBest optionfrom $0.33/hr (opens in new tab)
Compare GPUs →
or

How Much VRAM Does K2 Horizon MoVA 36B A4B Uncensored Need?

Select a quantization level; the listed hardware adheres to it.

QuantizationVRAM+32KOn RTX 4090 (reference)
16.5 GB20.2 GB2-bit · K-quant — 2-bit quantization with K-quant improvementsFits: 32K context · ~188 tok/s
18.8 GB22.6 GB3-bit medium — 3-bit medium quantizationFits: 32K context · ~175 tok/s
23.0 GB26.8 GB4-bit medium — 4-bit medium quantization — most popular sweet spotFits: 8K context · ~156 tok/s
27.2 GB31 GB5-bit medium — 5-bit medium quantization — good quality/size tradeoffDoes not fit: needs 29 GB
31.4 GB35.2 GB6-bit, very good — 6-bit quantization, very good qualityDoes not fit: needs 33.2 GB
38 GB41.8 GB8-bit, near-lossless — 8-bit quantization, near-losslessDoes not fit: needs 39.8 GB
75.4 GB79.2 GB16-bit · bfloat — Brain floating point 16 — preferred for trainingDoes not fit: needs 77.2 GB
est.calculated, no published file yet

Selected Q4_K_M

4-bit medium quantization

24.8GBat 16K context

Weights
22.5 GB
KV cache
2.0 GB
Overhead
0.3 GB

Fits RTX 4090 (reference) · fits up to 8K context · ~156 tok/s

≥50 tok/s needs ≥218 GB/s · RTX 5090 ~211, M3 Ultra ~72

Or rent RTX A6000 from $0.33/hr.

Which Hardware Can Run K2 Horizon MoVA 36B A4B Uncensored?

Q4_K_M · 23.0 GB

K2 Horizon MoVA 36B A4B Uncensored (Q4_K_M) requires 23.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 30+ GB is recommended. Using the full 524K context window can add up to 64.2 GB, bringing total usage to 87.2 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090. 100 devices with unified memory can run K2 Horizon MoVA 36B A4B Uncensored, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M5 Max (36 GB).

By VRAM tier · typical GPUs and Macs at Q4_K_M
8–16 GB · RTX 4060, RTX 3060 12GB, RTX 5060 Ti 16GBDoesn't fit — smallest quant needs 18.2 GB8 GB tier →12 GB tier →16 GB tier →
24 GB · RTX 4090referenceQ4_K_M · 24.8 GB · ~156 tok/s · 8Ktight24 GB tier →
Mac 32 GB · Mac Mini M4Q4_K_M · 24.8 GB · ~26 tok/s · 32KMacs →
Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 24.8 GB · ~62 tok/s · 256KMacs →

Alternatives to K2 Horizon MoVA 36B A4B Uncensored

Similar models

6
ModelScoreVRAMtok/sWhy it is similar
Qwen3.6 35B A3BAlibaba106.822 GB~151same size
MiniCPM V 4.7 35B A3BOpenBMB—22.1 GB~169same size
Ornith 1.5 35B A3Bornith-ai—22 GB~151same size
Agents A1InternScience—21.5 GB~171same size
Intern S2 PreviewInternLM—73.2 GB—needs > 24 GBsame size
Qwen3.5 35B A3B BaseAlibaba—22 GB~151same size

Specifications

Publisher
InfinimindCreations
Parameters
37.4B
Active parameters
4B per token
Architecture
K2HorizonForCausalLM
Context Length
524,288 tokens
Vocabulary Size
250,624
Release Date
2026-09-15
License
Apache 2.0

Resources

Frequently Asked Questions

How much VRAM does K2 Horizon MoVA 36B A4B Uncensored need?

K2 Horizon MoVA 36B A4B Uncensored requires 23.0 GB of VRAM at Q4_K_M, or 75.4 GB at BF16. Full 524K context adds up to 64.2 GB (87.2 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 37.4B × 4.8 bits ÷ 8 = 22.5 GB

KV Cache + Overhead ≈ 0.5 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 ≈ 64.7 GB (at full 524K context)

VRAM usage by quantization

23.0 GB
87.2 GB

Learn more about VRAM estimation →

Can NVIDIA GeForce RTX 4090 run K2 Horizon MoVA 36B A4B Uncensored?

Yes, at Q4_K_M (23.0 GB) or lower. Higher quantizations like Q5_K_M (27.2 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for K2 Horizon MoVA 36B A4B Uncensored?

For K2 Horizon MoVA 36B A4B Uncensored, Q4_K_M (23.0 GB) offers the best balance of quality and VRAM usage. Q5_K_M (27.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 16.5 GB.

VRAM requirement by quantization

Q2_K
16.5 GB
Q4_K_M ★
23.0 GB
Q5_K_M
27.2 GB
Q6_K
31.4 GB
Q8_0
38.0 GB
BF16
75.4 GB

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run K2 Horizon MoVA 36B A4B Uncensored on a Mac?

Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run K2 Horizon MoVA 36B A4B Uncensored, the smallest at Q2_K. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 23.0 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M1 Max (32 GB).

Can I run K2 Horizon MoVA 36B A4B Uncensored locally?

Yes — K2 Horizon MoVA 36B A4B Uncensored can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is K2 Horizon MoVA 36B A4B Uncensored?

At Q4_K_M, K2 Horizon MoVA 36B A4B Uncensored can reach ~99 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~156 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 → 2.5 GB active ÷ (8000 × 0.65) = 0.473 ms, plus 48 layers × 0.055 ms = 2.640 ms, so 1000 ÷ 3.113 ms = ~321 tok/s

Estimated speed at Q4_K_M (23.0 GB)

~321 tok/s
~156 tok/s
~321 tok/s
~293 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of K2 Horizon MoVA 36B A4B Uncensored?

At Q4_K_M, the download is about 22.47 GB. The full-precision BF16 version is 74.89 GB. The smallest option (Q2_K) is 15.91 GB.

Which GPUs can run K2 Horizon MoVA 36B A4B Uncensored?

7 consumer GPUs can run K2 Horizon MoVA 36B A4B Uncensored at Q4_K_M. Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090.

Which devices can run K2 Horizon MoVA 36B A4B Uncensored?

100 devices with unified memory can run K2 Horizon MoVA 36B A4B Uncensored at Q4_K_M (23.0 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.