Devstral Small vs GPT-5.4-Pro
Mistral Devstral Small | OpenAI GPT-5.4-Pro | |
|---|---|---|
| Overview | ||
| Company | Mistral | OpenAI |
| Release date | May 21 2025 | Mar 5 2026 |
| Access | Open Weight | Proprietary |
| Specifications | ||
Parameters | 24B | — |
| Benchmarks | ||
Web browsing BrowseCompCan the AI browse the web and track down hard-to-find answers? Higher is better. | — | 89.3% |
Multidisciplinary reasoning Humanity's Last Exam · with toolsHumanity'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. | — | 58.7% |
Abstract reasoning ARC-AGI-2Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. | — | 83.3% |
Advanced math FrontierMath · Tier 1–3Very hard, research-level math problems. Tiers 1–3 are the (still extremely difficult) lower tiers. Higher is better. | — | 50% |
Advanced math FrontierMath · Tier 4Very hard, research-level math problems. Tier 4 is the hardest — close to what professional research mathematicians tackle. Higher is better. | — | 38% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 94.4% |
Knowledge work GDPval (win/tie rate)How often the AI's work matches or beats a human expert's on real knowledge-work tasks. Higher is better. | — | 82% |
| Timeline | ||
| Release gap | Devstral Small shipped 288 days before GPT-5.4-Pro | |
Which is better: Devstral Small or GPT-5.4-Pro?
Devstral Small and GPT-5.4-Pro don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Devstral Small shipped 288 days before GPT-5.4-Pro, so benchmark comparisons should account for the intervening progress.
Devstral Small is open weight, while GPT-5.4-Pro is proprietary.
Direct benchmark comparisons are unavailable — Devstral Small and GPT-5.4-Pro don't publish scores on any of the same benchmarks.
Frequently asked questions
Devstral Small was released by Mistral on May 21 2025.
GPT-5.4-Pro was released by OpenAI on Mar 5 2026.
Devstral Small is an open weight model released by Mistral. GPT-5.4-Pro is a proprietary model released by OpenAI.
Other comparisons
Devstral Small vs Claude Sonnet 5GPT-5.4-Pro vs Claude Sonnet 5Devstral Small vs Gemini OmniGPT-5.4-Pro vs Gemini OmniDevstral Small vs Muse Spark 1.1GPT-5.4-Pro vs Muse Spark 1.1Devstral Small vs Grok 4.5GPT-5.4-Pro vs Grok 4.5Devstral Small vs DeepSeek-V4-ProGPT-5.4-Pro vs DeepSeek-V4-ProDevstral Small vs Kimi K2.7 CodeGPT-5.4-Pro vs Kimi K2.7 Code