Model
Model explorer

Gorilla 7B

OPEN
UC Berkeley & Microsoft Research · Gorilla family · released May 24, 2023

Zero-shot beats GPT-4 by 20.4pp and ChatGPT by 10.75pp on APIBench across TorchHub/HuggingFace/TensorHub.

ReasoningCodingVisionFunction callingTool useAgentic
1404.4
Elo · unrated
Parameters
7B
Active params
7B (dense)
Context
2K tokens
Architecture
LLaMA-7B dense Transformer, retrieval-aware instruction fine-tune for API calls
License
Apache 2.0 (delta weights; base requires Meta LLaMA research license)
Languages
—
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
APIBench (HuggingFace)Agents71.7%#1
best: this model · 71.7%
APIBench (TensorHub)Agents83.8%#1
best: this model · 83.8%
APIBench (TorchHub)Agents59.1%#1
best: this model · 59.1%
Run it locally
VRAM @ Q4
5 GB
VRAM @ FP16
14 GB
Fits on (Q4)
RTX 3060 12GBRTX 4070 Ti 16GBRTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB
Throughput data unavailable.
Quantizations
GGML · GGUF
Gorilla family
Elo progression across releases
Gorilla 7BMay 20231404.4
Resources
HFgorilla-llm/gorilla-7b-hf-delta-v1GHShishirPatil/gorilla
API price weights · each benchmark row carries its own source badge (see methodology)