LLM·Dex
Newsletter
ProprietaryOpenAIflagshiptextvisionaudio

GPT-5

OpenAI's unified flagship combining GPT-line breadth with built-in reasoning, replacing both GPT-4o and the o-series for most users.

Updated

Quick facts

Released
Aug 2025
Context
400K tokens
Output / 1M
$10.00 / 1M tokens
License
Proprietary

About GPT-5

GPT-5 launched in August 2025 as OpenAI's unified model, the first that ships with reasoning baked in rather than gated behind a separate o-series. Behind the scenes, a router decides per-request whether to spend extra reasoning tokens, which usually means latency and cost stay low for routine queries while harder questions get longer thinking.

For developers, GPT-5 brought the new Responses API, improved tool use, and a step-up in honesty / hallucination rates over GPT-4-class models. It became the default in ChatGPT for free, Plus, and Pro tiers shortly after launch.

GPT-5 is still the right choice for most production workloads in 2026 unless you specifically need GPT-5.5's mid-cycle improvements or a small-tier model for cost reasons.

Benchmarks

Published scores from OpenAI's model card or independent leaderboards. We do not publish numbers we cannot source, see methodology.

HumanEval
96.2
Python coding pass@1
MMLU
91.4
Broad academic knowledge
GPQA
89.4
Graduate-level reasoning
SWE-bench
74.9
Real software-engineering tasks
  • MMLU91.4
  • HumanEval96.2
  • GPQA89.4
  • SWE-bench Verified74.9

Capabilities

Strengths

  • Unified model, reasoning routed automatically per query
  • Excellent tool-use and JSON-mode discipline
  • Strong agent performance on SWE-bench Verified
  • Robust safety post-training reduces hallucinations vs. GPT-4 line

Tracked weaknesses

  • Reasoning routing means latency is unpredictable per query
  • Output cost is high relative to mid-tier alternatives

Pricing

Per-million-token rates as published by OpenAI.

TierPriceNotes
Input$1.25 / 1M tokensTokens you send to the model
Output$10.00 / 1M tokensTokens the model generates
Context400K tokensMax combined input + output

Call GPT-5 from your code

Drop-in snippet for the OpenAI SDK. Set your API key in the environment and run.

typescript
import OpenAI from "openai";

const client = new OpenAI({
  // Use OPENAI_API_KEY for OpenAI, or your provider's key + baseURL.
  apiKey: process.env.OPENAI_API_KEY,
});

const completion = await client.chat.completions.create({
  model: "gpt-5",
  messages: [
    { role: "user", content: "What's the time complexity of quicksort?" },
  ],
});

console.log(completion.choices[0].message.content);

Best for

Tasks where GPT-5 ranks among LLMDex's top picks.

Compare GPT-5 with…

Frequently asked

  • How much does GPT-5 cost per million tokens?
    GPT-5 is priced at $1.25 / 1M tokens for input tokens and $10.00 / 1M tokens for output tokens via the official OpenAI API at the time of writing.
  • What is GPT-5's context window?
    GPT-5 supports a context window of 400K tokens with up to 128K output tokens per response.
  • Is GPT-5 open source?
    No. GPT-5 is a closed-weight model, you can use it via OpenAI's API but the model weights are not publicly downloadable.
  • When was GPT-5 released?
    GPT-5 was released on Aug 7, 2025 by OpenAI.
  • What is GPT-5's knowledge cutoff?
    GPT-5's training data has a knowledge cutoff of Sep 2024. For information after that date you'll need a tool-use or web-search wrapper.
Friday digest

Intelligence, distilled weekly.

One short email every Friday, new model launches, leaderboard moves, and pricing drops. Curated by hand. Free, no spam.