WizardCoder 33B

WizardLM/WizardCoder-33B-V1.1

Parameters

33B

Context

16K

Downloads

104

Likes

136

Architecture

transformer

View on HuggingFace

GPU Requirements

VRAM / RAM / disk estimates by quantization and usage scenario

QuantizationMinimumRecommendedProduction
FP1616bit
82.7 GB
RAM 124.1 · Disk 134.9
102.0 GB
RAM 153.0 · Disk 160.9
202.4 GB
RAM 303.6 · Disk 199.9
FP88bit
43.8 GB
RAM 65.7 · Disk 86.9
61.4 GB
RAM 92.0 · Disk 112.9
160.0 GB
RAM 240.0 · Disk 151.9
INT88bit
43.8 GB
RAM 65.7 · Disk 86.9
61.4 GB
RAM 92.0 · Disk 112.9
160.0 GB
RAM 240.0 · Disk 151.9
AWQ4bit
25.1 GB
RAM 37.7 · Disk 63.9
41.8 GB
RAM 62.7 · Disk 89.9
139.6 GB
RAM 209.4 · Disk 128.9
GPTQ4bit
25.4 GB
RAM 38.0 · Disk 64.2
42.0 GB
RAM 63.1 · Disk 90.2
139.9 GB
RAM 209.8 · Disk 129.2
GGUF4bit
25.9 GB
RAM 38.8 · Disk 64.8
42.6 GB
RAM 63.8 · Disk 90.8
140.4 GB
RAM 210.6 · Disk 129.8

GPU Recommendations

Best ValueBEST

1× H100 SXM

80 GB VRAM

$1,315/mo

≈ 210.2 tok/s

Effective production throughput ≈ 147.1 tok/s

score 0.6258

Deploy on vast ↗
Cheapest

2× L4

48 GB VRAM

$152/mo

≈ 30.1 tok/s

Effective production throughput ≈ 21.1 tok/s

score 0.6156

Deploy on vast ↗
Performance

8× H200 SXM

1128 GB VRAM

$20,440/mo

≈ 963.6 tok/s

Effective production throughput ≈ 674.5 tok/s

score 0.2197

Min Complexity

1× A100 SXM4

80 GB VRAM

$225/mo

≈ 0 tok/s (N/A — benchmark unavailable)

Effective production throughput ≈ 0 tok/s

score 0.6246

Deploy on vast ↗

💬 Community — real deployment experience

0 Articles · 0 Benchmarks

View Community →

GPUs that fit (single card)

A100 SXM4

80 GB VRAM

$225

/mo · vast

RTX A6000

48 GB VRAM

$245

/mo · vast

L40S

48 GB VRAM

$313

/mo · vast

A100 PCIE

80 GB VRAM

$322

/mo · vast

RTX PRO 6000 Max-Q

48 GB VRAM

$779

/mo · vast

AmciHub

AI Model Deployment & Compute Intelligence — analyze AI model requirements, GPU performance, cloud pricing and deployment costs to find the right deployment solution.

© 2026 AmciHub. All rights reserved. Model → Requirement → GPU → Benchmark → Cloud → Cost → Recommendation