GPT-4 Vision Preview

Name: GPT-4 Vision Preview
Author: OpenAI

Released

2023-11-06

Last refreshed

2026-05-19

Status

Researched 52d ago

ProprietaryCommercial use: conditionalMultimodalCodingAgentsLong contextVision

GPT-4 Vision Preview is a released coding, agents, and long context model with 128k context; evaluate it while provider pricing coverage matures.

Use it for

Teams evaluating coding, agents, and long context
Workloads that can use a 128k context window

Do not use it for

Cost-sensitive launches that need sourced token pricing
Strict JSON or tool-calling flows

Specifications

Family: GPT-4
Released: 2023-11-06
Context: 128k
Parameters: 1.76T (8x222B MoE)*
Architecture: Mixture of Experts
Knowledge cutoff: 2023-04
Specialization: general
Openness: Proprietary
License: ProprietaryCommercial use: conditional
Weights: Not released
Code: Unknown
Training: Fine-tuned

Created by

OpenAI

Cutting-edge research and development.

San Francisco, California, United States

Founded 2015

Website

Pricing

No tracked provider token pricing is available yet.

Providers(1)

Azure OpenAI

View 1 provider route

About

GPT-4 Vision Preview is OpenAI's GPT-4 model with multimodal text and image input. It is deprecated (originally released 2023-11-06); use it only for reproducing earlier results or evaluating drift over time.

GPT-4 Vision Preview is a proprietary model in the GPT-4 family. The structured metadata tracks a 128k-token context window, multimodal input, and code execution. This page tracks provider routes through Azure OpenAI, with the cheapest tracked route listed at $10 input and $40 output per 1M tokens. Headline tracked benchmarks include Massive Multi-discipline Multimodal Understanding 56.0 and GAOKAO-MM 48.1.

Top use-case fit: coding, agents, and build tasks

Coding

Included by capability and metadata signals in the decision map.

Agents

Included by capability and metadata signals in the decision map.

Long context

Included by capability and metadata signals in the decision map.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Available via routers & gateways(5)

LiteLLM

Gateway

Open-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.

Free OSSAzure OpenAI

Portkey

Gateway

Production AI gateway routing to 1,600+ LLMs with failover, load balancing, semantic caching, and guardrails; Apache 2.0 core is fully self-hostable with the complete feature set.

SubscriptionAzure OpenAI

Azure AI Foundry Model Router

Router

Microsoft Azure AI Foundry's native model router that uses a trained ML model to route each prompt in real time to the optimal Azure-hosted model, with Balanced/Cost/Quality mode selection and automatic failover.

PassthroughAzure OpenAI

Helicone

Gateway

Observability-first AI gateway with routing, caching, rate limiting, and request tracing; Apache 2.0 open-source core with a managed hosted tier for logging and analytics.

SubscriptionAzure OpenAI

Kong AI Gateway

Gateway

Multi-LLM AI gateway built on Kong Gateway 3.x, adding semantic routing, load balancing, guardrails, and MCP traffic analytics as plugins over Kong's existing API management platform.

SubscriptionAzure OpenAI

Capabilities

VisionMultimodalCode Execution

Benchmark peer barsfor Vision

Massive Multi-discipline Multimodal UnderstandingRank 37 of 46

Qwen3.6-Plus

86.0

ByteDance Doubao Seed 2.0 Pro

85.4

Qwen3.5-397B-A17B

85.0

Gemini 3.5 Flash

83.6

GPT-4 Vision Previewcurrent

56.0

GAOKAO-MMRank 1 of 7

GPT-4 Vision Previewcurrent

48.1

Qwen-VL-Plus

41.2

Gemini 1.0 Pro Vision

35.1

InternLM XComposer2 VL 7B

33.2

LLaVA 1.5 13B

18.0

Benchmark scores(2)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.