GPT-5.4 minivsGLM-5
GPT-5.4 mini | GLM-5 | |
|---|---|---|
| Specifications | ||
ParametersA rough measure of how big the model is. More parameters usually means more capable and more expensive to run, though it is a poor guide on its own — a smaller, newer model often beats a larger, older one. | — | 744B |
Context windowHow much text the model can hold in mind at once — your question, any documents you attach, the conversation so far, and its own reply. Go past it and the earliest part falls out of view. | 400k | — |
| API pricingUSD per 1M tokens · lower wins | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.75 | $1.00 |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $4.50 | $3.20 |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | $0.075 | $0.20 |
Cheapest inputLowest input rate across third-party providers, excluding the lab itself. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.75Azure | $0.60DeepInfra |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | $4.50Azure | $1.92StreamLake |
| Benchmarks | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 32% | 28% |
| BenchmarksPublished by one model only | ||
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | — | 77.8% |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | — | 73.3% |
Supabase Evals · with skillsSupabase coding — Supabase's own open benchmark: a coding agent is dropped into a real Supabase project and asked to do real work — set up a schema, fix a broken security policy, debug an Edge Function — and every run is checked against a live Supabase stack. This is the headline number, where the agent has Supabase's own skills loaded, as most people building on Supabase would. The score is the share of scenarios it got right. Higher is better. | 81.8% | — |
Supabase Evals · no skillsSupabase coding — The same Supabase scenarios, but with none of Supabase's skills loaded — so it measures what the model already knows about building on Supabase, rather than how well it follows Supabase's supplied instructions. Higher is better. | 63.6% | — |
Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | — | 56.2% |
BU BenchBrowser agent — Can the AI drive a real web browser to finish tasks — clicking, filling forms, and navigating sites the way a person would? Run by Browser Use on their BU Bench task set. Higher is better. | 36% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | — | 75.9% |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | — | 50.4% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 86% |
| Overview | ||
| Company | OpenAI | Z.ai |
| Release date | Mar 17 2026 | Feb 12 2026 |
| Access | Proprietary | Open Weight |
Other comparisons
GPT-5.4 minivsClaude Opus 5GLM-5vsClaude Opus 5GPT-5.4 minivsGemini 3.7 FlashGLM-5vsGemini 3.7 FlashGPT-5.4 minivsMuse GlimmerGLM-5vsMuse GlimmerGPT-5.4 minivsGrok 4.6GLM-5vsGrok 4.6GPT-5.4 minivsDeepSeek-V4-Pro-0813GLM-5vsDeepSeek-V4-Pro-0813GPT-5.4 minivsMistral Medium 3.5GLM-5vsMistral Medium 3.5
Frequently asked questions
GPT-5.4 mini leads GLM-5 on 1 of the 1 benchmark they both report (BullshitBench v2). GPT-5.4 mini is cheaper on input: $0.75 vs $1.00 per million tokens. GLM-5 is cheaper on output: $3.20 vs $4.50 per million tokens. GLM-5 shipped 33 days before GPT-5.4 mini, so benchmark comparisons should account for the intervening progress.
GPT-5.4 mini is proprietary, while GLM-5 is open weight.
On BullshitBench v2, GPT-5.4 mini leads at 32% vs GLM-5 at 28%.
GPT-5.4 mini was released by OpenAI on Mar 17 2026.
GLM-5 was released by Z.ai on Feb 12 2026.
GPT-5.4 mini is cheaper on input: $0.75 vs $1.00 per million tokens. GLM-5 is cheaper on output: $3.20 vs $4.50 per million tokens. Rates are pay-as-you-go API prices verified on August 18, 2026.
GPT-5.4 mini is a proprietary model released by OpenAI. GLM-5 is an open weight model released by Z.ai.