Model
Model explorer

WizardCoder-Python-34B V1.0

OPEN
Microsoft · WizardCoder family · released Aug 26, 2023

Code Evol-Instruct fine-tune of Code Llama-Python 34B; 73.2 HumanEval pass@1, top open coding model of 2023 (ICLR 2024 paper).

ReasoningCodingVisionFunction callingTool useAgentic
745.4
Elo · unrated
Parameters
34B
Active params
34B (dense)
Context
16K tokens
Architecture
Dense decoder-only Transformer (Code Llama-Python 34B backbone)
License
Llama 2 Community License
Languages
—
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
HumanEval+Coding64.6%#23
best: Mistral Small 3.2 (24B) · 92.9%
HumanEvalCoding73.2%#73
best: Claude Opus 4.5 · 99.4%
MBPP+Coding59.9%#30
best: Llama 3.1 405B · 88.6%
MBPPCoding73.2%#28
best: Llama-3.3-Nemotron-Super-49B v1 (Reasoning On) · 91.3%
Run it locally
VRAM @ Q4
20 GB
VRAM @ FP16
68 GB
Fits on (Q4)
RTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB
Throughput data unavailable.
Quantizations
GGUF Q4_K_M · GPTQ · AWQ
WizardCoder family
Elo progression across releases
WizardCoder-Python-34B V1.0Aug 2023745.4
Resources
HFWizardLMTeam/WizardCoder-Python-34B-V1.0GHnlpxucan/WizardLM
API price weights · each benchmark row carries its own source badge (see methodology)