What are the hardware requirements for SqlCoder Qwen2.5 8bit?

SqlCoder Qwen2.5 8bit requires approximately 0.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed SqlCoder Qwen2.5 8bit and how large is it?

SqlCoder Qwen2.5 8bit is developed by imsanjoykb, a model with 32b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does SqlCoder Qwen2.5 8bit perform on standard benchmarks?

SqlCoder Qwen2.5 8bit has the following published scores: HumanEval (Python) 83.2, LMSYS ELO 1217. Compare against reference models on this page or on the LLM Explorer leaderboards.

SqlCoder Qwen2.5 8bit by imsanjoykb — VRAM 0.3GB

Name: SqlCoder Qwen2.5 8bit
Author: imsanjoykb

SqlCoder Qwen2.5 8bit is an open-source language model by imsanjoykb. Features: 32b LLM, VRAM: 0.3GB, License: apache-2.0, Quantized, Instruction-Based, Code Generating, LLM Explorer Score: 0.27, ELO: 1217, HumanEval: 83.2.

8bit Adapter Adapter-transformers Base model:adapter:qwen/qwen2.... Base model:qwen/qwen2.5-coder-... Codegen Conversational Doi:10.57967/hf/4444 En Finetuned Inference-endpoint, Instruct Lora Pytorch, Quantized Region:us Safetensors Sql-code-generation, Unsloth,

Model Card on HF 🤗: https://huggingface.co/imsanjoykb/sqlCoder-Qwen2.5-8bit

SqlCoder Qwen2.5 8bit Benchmarks

LMSys ELO 1217 vs 1272 (so35) -4.3%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

SqlCoder Qwen2.5 8bit Parameters and Internals

LLM Name	SqlCoder Qwen2.5 8bit
Repository 🤗	https://huggingface.co/imsanjoykb/sqlCoder-Qwen2.5-8bit
Base Model(s)	Qwen/Qwen2.5-Coder-32B-Instruct Qwen/Qwen2.5-Coder-32B-Instruct
Model Size	32b
Required VRAM	0.3 GB
Updated	2026-08-01
Maintainer	imsanjoykb
Instruction-Based	Yes
Model Files	0.3 GB
Supported Languages	en
Quantization Type	8bit
Generates Code	Yes
Model Architecture	Adapter
License	apache-2.0
Model Max Length	32768
Is Biased	none
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|PAD_TOKEN\|>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	o_proj\|v_proj\|down_proj\|up_proj\|q_proj\|gate_proj\|k_proj
LoRA Alpha	16
LoRA Dropout	0
R Param	16
Errors	replace

Rank the SqlCoder Qwen2.5 8bit 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