GPT-5 vs GPT-5.4 Mini
At a Glance
| Compare |  GPT-5OpenAI | GPT-5.4 MiniOpenAI |
|---|---|---|
| Intelligence, Cost, and Efficiency | ||
| IntelligenceHigher is better · MM Intelligence v2.5 | #41 of 4615.0 score · 2/3 sources · provisional · missing LiveBench · full-core range 10.0–43.3 | #38 of 4620.2 score · 3/3 sources · complete |
| CostLower is better · Published-token output estimate | UnrankedNot in the 44-model eligible cohort | #32 of 44$0.214 per LiveBench case |
| EfficiencyHigher is better · MM Efficiency v1.5 | UnrankedNot in the 38-model eligible cohort | #38 of 3826.5 score · 3/3 sources · complete |
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.625Openrouter ↗ · Sep 3, 2026 | $0.75Openai ↗ · Sep 3, 2026 |
| Output priceFrom · USD / 1M tokens | $5.00Openrouter ↗ · Sep 3, 2026 | $4.50Openai ↗ · Sep 3, 2026 |
| Context windowMaximum documented tokens | 400K | 400K |
| Model facts checked | Sep 3, 2026View model evidence → | Sep 3, 2026View model evidence → |
Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →
Available Benchmarks
| Benchmark | GPT-5 | GPT-5.4 Mini |
|---|---|---|
| ARC-AGI-1verified-v1-ba05d69f6453 · verified_score · leader | 65.67100% of row best · percent · GPT-5 (High) | 63.6797% of row best · percent · GPT-5.4 Mini (XHigh) |
| ARC-AGI-2verified-v2-6c676fa3e9af · verified_score · leader | 9.8652% of row best · percent · GPT-5 (High) | 18.90100% of row best · percent · GPT-5.4 Mini (XHigh) |
| LMArena Text Arenatext-2026-09-13-d25aabda0010 · arena_rating · statistical tie | 1,406.06100% of row best · rating · gpt-5-high; 95% CI [1401.61960044, 1410.50585480]; votes 31428; rank 136 | 1,412.11100% of row best · rating · gpt-5.4-mini-high; 95% CI [1408.29499732, 1415.92580718]; votes 59387; rank 127 |
| LMArena Vision Arenavision-2026-09-13-d25aabda0010 · arena_rating · leader | 1,209.2297% of row best · rating · gpt-5-high; 95% CI [1202.34954059, 1216.08580885]; votes 34351; rank 76 | 1,245.09100% of row best · rating · gpt-5.4-mini-high; 95% CI [1238.67164332, 1251.50821847]; votes 24041; rank 59 |
| ToneBench2026-09-11-10-task-4ef099199c9c · overall_score · unverified | 77.8295% of row best · points · GPT-5 · 5,896 output tokens / case | 82.16100% of row best · points · GPT-5.4 mini (high) · 9,924 output tokens / case |
| Overall ResultCounted from the protocol-matched rows above · 1 tie | 1 benchmark win | 2 benchmark winsOverall lead |
Third-party benchmark Only like-for-like primary-publisher results are shown; raw scores, relative scores, configuration, and token spend remain visible.
Side-by-Side Facts
| Field | GPT-5 | GPT-5.4 Mini |
|---|---|---|
| Developer | OpenAI | OpenAI |
| Family | Gpt 5 | Gpt 5 4 |
| Model | GPT-5 | GPT-5.4 Mini |
| Version | GPT-5 | GPT-5.4 Mini |
| Lifecycle | active | active |
| Released | 2025-08-07 | 2026-03-17 |
| Knowledge cutoff | 2024-09-30 | 2025-08-31 |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Context window | 400K | 400K |
| Total parameters | Unknown | Unknown |
| Active parameters | Unknown | Unknown |
| License | Unknown | Unknown |
| Open weights | No | No |
| API available | Yes | Yes |
| Self-hostable | No | No |
| Provider access | Openai (Standard), Openrouter (Standard) | Openai (Standard), Openrouter (Standard) |
| Capabilities | chat, generation, reasoning, structured_outputs, tools | chat, generation, reasoning, structured_outputs, tools |
GPT-5 Capabilities
GPT-5.4 Mini Capabilities
Related Comparisons
All image comparisons →| A | Pair | B | Context |
|---|---|---|---|
GPT-5OpenAI | vs | GPT-6 AstraOpenAI | family variantsimage, text |
Claude Fable 5.1Anthropic | vs | GPT-5OpenAI | cross-developer peersimage, text |
 Gemini 3.1 ProGoogle DeepMind | vs | GPT-5OpenAI | cross-developer peerstext |
 DeepSeek-V4.1-FlashDeepSeek | vs | GPT-5OpenAI | cross-developer peersimage, text |
GPT-5OpenAI | vs | Grok 4.6xAI | cross-developer peersimage, text |
GPT-5OpenAI | vs |  Qwen3.8-MaxQwen | cross-developer peersimage, text |
 Kimi-K3Moonshot AI | vs | GPT-5OpenAI | cross-developer peersimage, text |
 MiniMax-M3MiniMax | vs | GPT-5OpenAI | cross-developer peersimage, text |
GPT-5OpenAI | vs |  GLM-5.3Z.ai | cross-developer peerstext |
GPT-5OpenAI | vs |  Hy4 previewTencent | cross-developer peerstext |
Seed 2.1 ProByteDance Seed | vs | GPT-5OpenAI | cross-developer peersimage, text |
 Mistral Large 3Mistral AI | vs | GPT-5OpenAI | cross-developer peersimage, text |
Primary Evidence
Sources and Freshness
Questions
GPT-5 vs GPT-5.4 Mini FAQs
Is GPT-5 or GPT-5.4 Mini better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both GPT-5 and GPT-5.4 Mini, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, GPT-5 or GPT-5.4 Mini?+
GPT-5 is $0.625 and GPT-5.4 Mini is $0.75 per million tokens, so GPT-5 is cheaper on this metric. GPT-5 is $5.00 and GPT-5.4 Mini is $4.50 per million tokens, so GPT-5.4 Mini is cheaper on this metric.
Which has a larger context window, GPT-5 or GPT-5.4 Mini?+
Neither model has a larger sourced context window in this comparison. GPT-5 is 400K and GPT-5.4 Mini is 400K.
Which performs better in benchmarks, GPT-5 or GPT-5.4 Mini?+
GPT-5.4 Mini leads the current overall benchmark count. The result uses 4 protocol-matched benchmarks from 2 publishers; it is not a universal quality score.
Can GPT-5 or GPT-5.4 Mini be self-hosted?+
Both models have the same recorded self-hosting status: unsupported. GPT-5 is not marked open weight; GPT-5.4 Mini is not marked open weight.
Can GPT-5 and GPT-5.4 Mini understand images?+
GPT-5 is documented with image input; GPT-5.4 Mini is documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, GPT-5 or GPT-5.4 Mini?+
Neither has a larger sourced maximum output. GPT-5 is 128K and GPT-5.4 Mini is 128K.
Do GPT-5 and GPT-5.4 Mini support reasoning and tool use?+
GPT-5: reasoning, tool calling, and image input. GPT-5.4 Mini: reasoning, tool calling, and image input. Feature support does not establish relative quality.
Which is available from more inference providers, GPT-5 or GPT-5.4 Mini?+
GPT-5 has 2 sourced provider routes; GPT-5.4 Mini has 2, a tie.
Which offers better value, GPT-5 or GPT-5.4 Mini?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.