OpenGVLab: InternVL3 78B

Image input Text input Text output
Author's Description

The InternVL3 series is an advanced multimodal large language model (MLLM). Compared to InternVL 2.5, InternVL3 demonstrates stronger multimodal perception and reasoning capabilities. In addition, InternVL3 is benchmarked against the Qwen2.5 Chat models, whose pre-trained base models serve as the initialization for its language component. Benefiting from Native Multimodal Pre-Training, the InternVL3 series surpasses the Qwen2.5 series in overall text performance.

Key Specifications
Cost
$
Context
32K
Parameters
78B
Released
Sep 15, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Top Logprobs Stop Structured Outputs Logprobs Presence Penalty Frequency Penalty Top P Max Tokens Min P Response Format Logit Bias Seed Temperature
Features

This model supports the following features:

Response Format Structured Outputs
Performance Summary

The OpenGVLab: InternVL3 78B model demonstrates moderate speed performance, ranking in the 38th percentile across benchmarks. However, it excels in cost-efficiency, consistently offering among the most competitive pricing, placing it in the 92nd percentile. Notably, the model exhibits exceptional reliability with a perfect 100% success rate across all benchmarks, indicating minimal technical failures. InternVL3 78B shows strong performance in knowledge-based tasks, achieving perfect accuracy in General Knowledge and Ethics, often being the most accurate model at its price point and among models of similar speed. It also performs very well in Hallucinations (98.0% accuracy), effectively acknowledging uncertainty. While its Email Classification accuracy is respectable at 97.0%, its performance in Instruction Following (57.6%) and Reasoning (60.0%) is more moderate. Coding performance is solid at 84.0%. Its key strengths lie in its high accuracy for factual recall and ethical reasoning, coupled with outstanding reliability and cost-effectiveness. Its primary area for improvement appears to be in complex multi-step instruction following and abstract reasoning tasks.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.07
Completion $0.26

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Chutes
Chutes | opengvlab/internvl3-78b 32K $0.07 / 1M tokens $0.26 / 1M tokens
Chutes
Chutes | opengvlab/internvl3-78b 32K $0.07 / 1M tokens $0.26 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by opengvlab