llmrun
zgcagiZgcmForCausalLM

ZGCM 1 7B — Hardware Requirements & GPU Compatibility

ChatMathReasoning

ZGCM 1 7B is a 7.4B-parameter open language model from zgcagi. It supports a context window of up to 262,144 tokens. At BF16 it needs about 15.36 GB of VRAM — see which GPUs and Macs can run it below.

7.4B params · dense262K contextMIT2.2K downloads35 likes
A

Runs well on a 24 GB GPU

Runs from16 GB BF16 · 4KBest on24 GB BF16 · ~43 tok/s · 64KMacs from24 GB BF16 · ~14 tok/s

MIT

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run ZGCM 1 7B

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 zgcagi/ZGCM-1-7B --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 3090 (24 GB) (6 GB headroom) · BF16

Generation speed
~40 tok/s
generation speed
Cost per 1M output tokens
$1.39
per 1M output tokens
Deployon Vast.aiBest valuefrom $0.20/hr (opens in new tab)
Compare GPUs →
or

How Much VRAM Does ZGCM 1 7B Need?

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

QuantizationVRAM+32KOn RTX 4090 (reference)
15.4 GB19.4 GB16-bit · bfloat — Brain floating point 16 — preferred for trainingFits: 64K context · ~43 tok/s
est.calculated, no published file yet

Selected BF16

Brain floating point 16

17.2GBat 16K context

Weights
14.8 GB
KV cache
2.2 GB
Overhead
0.3 GB

Fits RTX 4090 (reference) · 6.8 GB free · ~43 tok/s

≥50 tok/s needs ≥1.18 TB/s · RTX 5090 ~76

Which Hardware Can Run ZGCM 1 7B?

BF16 · 15.4 GB

ZGCM 1 7B (BF16) requires 15.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 20+ GB is recommended. Using the full 262K context window can add up to 34.1 GB, bringing total usage to 49.5 GB. 26 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 5080. 122 devices with unified memory can run ZGCM 1 7B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M5 Pro (24 GB).

By VRAM tier · typical GPUs and Macs at BF16
8–12 GB · RTX 4060, RTX 3060 12GBDoesn't fit — smallest quant needs 17.2 GB8 GB tier →12 GB tier →
16 GB · RTX 5060 Ti 16GBBF16 · 17.2 GB · ~19 tok/s · 4Ktight16 GB tier →
24 GB · RTX 4090referenceBF16 · 17.2 GB · ~43 tok/s · 64K24 GB tier →
Mac 32 GB · Mac Mini M4BF16 · 17.2 GB · ~6 tok/s · 64KMacs →
Mac 64 GB · Mac Studio M4 MaxBF16 · 17.2 GB · ~25 tok/s · 256KMacs →

Alternatives to ZGCM 1 7B

Similar models

6
ModelScoreVRAMtok/sWhy it is similar
Ling 3.0 TinyInclusion AI—5.5 GB~364same size
Olmo 3 7B InstructAllen AI—8.4 GB~114same size
Phi Mini MoE InstructMicrosoft—5.4 GB~239same size
Gemma 3n E4B ITGoogle—7.7 GB~127same size
Falcon H1 7B InstructTII UAE—6.3 GB~130same size
DeepSeek R1 Distill Qwen 7BDeepSeek—5.8 GB~131same size

Specifications

Publisher
zgcagi
Parameters
7.4B
Architecture
ZgcmForCausalLM
Context Length
262,144 tokens
Vocabulary Size
155,136
Release Date
2026-09-07
License
MIT

Resources

Frequently Asked Questions

How much VRAM does ZGCM 1 7B need?

ZGCM 1 7B requires 15.4 GB of VRAM at BF16. Full 262K context adds up to 34.1 GB (49.5 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 7.4B × 16 bits ÷ 8 = 14.8 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 ≈ 34.7 GB (at full 262K context)

VRAM usage by quantization

15.4 GB
49.5 GB

Learn more about VRAM estimation →

Can I run ZGCM 1 7B on a Mac?

Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run ZGCM 1 7B. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 15.4 GB of usable unified memory (RAM minus macOS overhead).

Can I run ZGCM 1 7B locally?

Yes — ZGCM 1 7B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is ZGCM 1 7B?

At BF16, ZGCM 1 7B can reach ~313 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~43 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 ÷ 15.36 × 0.65 = ~339 tok/s

Estimated speed at BF16 (15.4 GB)

~339 tok/s
~43 tok/s
~339 tok/s
~313 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 ZGCM 1 7B?

At BF16, the download is about 14.79 GB.

Which GPUs can run ZGCM 1 7B?

26 consumer GPUs can run ZGCM 1 7B at BF16. Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3090 Ti, AMD Radeon RX 6800. 7 GPUs have plenty of headroom for comfortable inference.

Which devices can run ZGCM 1 7B?

122 devices with unified memory can run ZGCM 1 7B at BF16 (15.4 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.