llmrun
Apexintelligence-AIQwen3_5MoeForConditionalGeneration

Apex 35B v1.0 — Hardware Requirements & GPU Compatibility

ChatVisionFunctions

Apex 35B v1.0 is a 36.0B-parameter open language model from Apexintelligence-AI. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 21.95 GB of VRAM — see which GPUs and Macs can run it below.

36.0B params · MoE4.7B active262K contextApache 2.01.5K downloads66 likes
Based on Qwen3.6 35B A3B
A

Runs well on a 24 GB GPU

Runs from16 GB Q2_K · 32KBest on24 GB Q4_K_M · ~151 tok/s · 128KMacs from24 GB Q3_K_M · ~53 tok/s

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Apex 35B v1.0

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

vLLM
one command · needs vLLM installed
$vllm serve Apexintelligence-AI/Apex-35B-V1.0 --max-model-len 16384

vLLM repo with its own quantisation

Download: Weights on Hugging Face

install, OS, other quants

Ways to runOther ways

each opens step-by-step

Run in cloud

Runs on RTX A6000 (48 GB) (26 GB headroom) · Q4_K_M

Generation speed
~125 tok/s
generation speed
Cost per 1M output tokens
$0.73
per 1M output tokens
Deployon RunPodBest valuefrom $0.33/hr (opens in new tab)

$5+ bonus for new users

Compare GPUs →
or

How Much VRAM Does Apex 35B v1.0 Need?

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

QuantizationVRAM+32KOn RTX 4090 (reference)
15.7 GB15.9 GB2-bit · K-quant — 2-bit quantization with K-quant improvementsFits: 256K context · ~187 tok/s
17.9 GB18.2 GB3-bit medium — 3-bit medium quantizationFits: 256K context · ~172 tok/s
21.9 GB22.2 GB4-bit medium — 4-bit medium quantization — most popular sweet spotFits: 128K context · ~151 tok/s
26 GB26.3 GB5-bit medium — 5-bit medium quantization — good quality/size tradeoffDoes not fit: needs 26.1 GB
30.0 GB30.3 GB6-bit, very good — 6-bit quantization, very good qualityDoes not fit: needs 30.1 GB
36.3 GB36.6 GB8-bit, near-lossless — 8-bit quantization, near-losslessDoes not fit: needs 36.4 GB
72.3 GB72.5 GB16-bit · bfloat — Brain floating point 16 — preferred for trainingDoes not fit: needs 72.4 GB
est.calculated, no published file yet

Selected Q4_K_M

4-bit medium quantization

22.0GBat 16K context

Weights
21.6 GB
KV cache
0.2 GB
Overhead
0.3 GB

Fits RTX 4090 (reference) · 2.0 GB free · ~151 tok/s

≥50 tok/s needs ≥250 GB/s · RTX 5090 ~213, RTX 3090 Ti ~151

Or .

Which Hardware Can Run Apex 35B v1.0?

Q4_K_M · 21.9 GB

Apex 35B v1.0 (Q4_K_M) requires 21.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 29+ GB is recommended. Using the full 262K context window can add up to 2.6 GB, bringing total usage to 24.6 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090. 100 devices with unified memory can run Apex 35B v1.0, 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
8–12 GB · RTX 4060, RTX 3060 12GBDoesn't fit — smallest quant needs 15.7 GB8 GB tier →12 GB tier →
16 GB · RTX 5060 Ti 16GBQ2_K · 15.7 GB · ~108 tok/s · 32K16 GB tier →
24 GB · RTX 4090referenceQ4_K_M · 22 GB · ~151 tok/s · 128K24 GB tier →
Mac 32 GB · Mac Mini M4Q4_K_M · 22 GB · ~24 tok/s · 256KMacs →
Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 22 GB · ~64 tok/s · 256KMacs →

Alternatives to Apex 35B v1.0

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
EXAONE 4.5 33BLGAI-EXAONE—23.9 GB~33same size

Specifications

Publisher
Apexintelligence-AI
Parameters
36.0B
Active parameters
4.7B per token
Architecture
Qwen3_5MoeForConditionalGeneration
Context Length
262,144 tokens
Vocabulary Size
248,320
Release Date
2026-10-28
License
Apache 2.0

Resources

Frequently Asked Questions

How much VRAM does Apex 35B v1.0 need?

Apex 35B v1.0 requires 21.9 GB of VRAM at Q4_K_M, or 72.3 GB at BF16. Full 262K context adds up to 2.6 GB (24.6 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 36.0B × 4.8 bits ÷ 8 = 21.6 GB

KV Cache + Overhead ≈ 0.3 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 ≈ 2.9 GB (at full 262K context)

VRAM usage by quantization

21.9 GB
24.6 GB

Learn more about VRAM estimation →

Can NVIDIA GeForce RTX 4090 run Apex 35B v1.0?

Yes, at Q4_K_M (21.9 GB) or lower. Higher quantizations like Q5_K_M (26 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for Apex 35B v1.0?

For Apex 35B v1.0, Q4_K_M (21.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (26 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 15.7 GB.

VRAM requirement by quantization

Q2_K
15.7 GB
Q4_K_M ★
21.9 GB
Q5_K_M
26.0 GB
Q6_K
30.0 GB
Q8_0
36.3 GB
BF16
72.3 GB

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Apex 35B v1.0 on a Mac?

Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Apex 35B v1.0, the smallest at Q3_K_M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 21.9 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M1 Max (32 GB).

Can I run Apex 35B v1.0 locally?

Yes — Apex 35B v1.0 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Apex 35B v1.0?

At Q4_K_M, Apex 35B v1.0 can reach ~116 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~151 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 → 2.9 GB active ÷ (8000 × 0.65) = 0.557 ms, plus 40 layers × 0.055 ms = 2.200 ms, so 1000 ÷ 2.757 ms = ~363 tok/s

Estimated speed at Q4_K_M (21.9 GB)

~363 tok/s
~151 tok/s
~363 tok/s
~322 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 Apex 35B v1.0?

At Q4_K_M, the download is about 21.57 GB. The full-precision BF16 version is 71.90 GB. The smallest option (Q2_K) is 15.28 GB.

Which GPUs can run Apex 35B v1.0?

7 consumer GPUs can run Apex 35B v1.0 at Q4_K_M. Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090.

Which devices can run Apex 35B v1.0?

100 devices with unified memory can run Apex 35B v1.0 at Q4_K_M (21.9 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.