What are the hardware requirements for Phind CodeLlama 34B V2?

Phind CodeLlama 34B V2 requires approximately 67.5 GB of VRAM and supports a context window of 16K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Phind CodeLlama 34B V2 and how large is it?

Phind CodeLlama 34B V2 is developed by phind, a model with 34b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Phind CodeLlama 34B V2 perform on standard benchmarks?

Phind CodeLlama 34B V2 has the following published scores: MMLU 25.76. Compare against reference models on this page or on the LLM Explorer leaderboards.

Phind CodeLlama 34B V2 by phind — VRAM 67.5GB, 16K context

Name: Phind CodeLlama 34B V2
Author: phind

Phind CodeLlama 34B V2 is an open-source language model by phind. Features: 34b LLM, VRAM: 67.5GB, Context: 16K, License: llama2, Code Generating, LLM Explorer Score: 0.15, Arc: 24.6, HellaSwag: 27.6, MMLU: 25.8, GSM8K: 23.2.

Code llama Codegen Endpoints compatible Llama Model-index Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/Phind/Phind-CodeLlama-34B-v2

Phind CodeLlama 34B V2 Parameters and Internals

Model Type

text generation, multi-lingual

Additional Notes

This model is instruction-tuned to be steerable and easy-to-use. More information can be found in the blog post.

Supported Languages

Python (proficient), C/C++ (proficient), TypeScript (proficient), Java (proficient), multi-lingual (general proficiency)

Training Details

Data Sources:

proprietary dataset of programming problems and solutions

Data Volume:

1.5B tokens

Methodology:

instruction-tuned (Alpaca/Vicuna format)

Context Length:

4096

Training Time:

15 hours

Hardware Used:

32x A100-80GB GPUs

Model Architecture:

DeepSpeed ZeRO 3 with Flash Attention 2

Input Output

Input Format:

Alpaca/Vicuna instruction format

Accepted Modalities:

text

Output Format:

text

LLM Name	Phind CodeLlama 34B V2
Repository 🤗	https://huggingface.co/Phind/Phind-CodeLlama-34B-v2
Model Size	34b
Required VRAM	67.5 GB
Updated	2026-07-21
Maintainer	phind
Model Type	llama
Model Files	9.8 GB: 1-of-7 9.7 GB: 2-of-7 9.7 GB: 3-of-7 9.7 GB: 4-of-7 9.7 GB: 5-of-7 9.7 GB: 6-of-7 9.2 GB: 7-of-7
Generates Code	Yes
Model Architecture	LlamaForCausalLM
License	llama2
Context Length	16384
Model Max Length	16384
Transformers Version	4.33.0.dev0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Phind CodeLlama 34B V2 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Quantized Models of the Phind CodeLlama 34B V2

Model	Likes	Downloads	VRAM
Phind Codellama 34B V2 EXL2	16	8	0 GB
Phind CodeLlama 34B V2 GGUF	171	3383	14 GB
Phind CodeLlama 34B V2 AWQ	35	42	18 GB
Phind CodeLlama 34B V2 GPTQ	90	64	17 GB

Best Alternatives to Phind CodeLlama 34B V2

Best Alternatives	Context / RAM	Downloads	Likes
...gpt 32K Codellama 34B Instruct	32K / 67.5 GB	56	2
CodeLlama 34B Instruct Hf	16K / 67.5 GB	22872	304
CodeLlama 34B Instruct Hf	16K / 1.4 GB	7	3
CodeLlama 34B Instruct Hf	16K / 1.4 GB	5	3
ReflectionCoder CL 34B	16K / 67.6 GB	7735	0
Speechless Codellama 34B V2.0	16K / 67.5 GB	96	17
CodeLlama 34B Hf	16K / 67.5 GB	27673	178
Phind CodeLlama 34B V1	16K / 67.5 GB	165	320
Phind CodeLlama 34B Python V1	16K / 67.5 GB	180	252
CodeLlama 34B Python Hf	16K / 67.5 GB	459	99

Note: green Score (e.g. "73.2") means that the model is better than Phind/Phind-CodeLlama-34B-v2.