Compare / GPT-4.1 nano vs GPT-5.4 nano

GPT-4.1 nano vs GPT-5.4 nano

Which is cheaper, which is more capable, and which does better at each kind of work, from 17 results on 3 sources that measured both models.

OpenAIGPT-4.1 nano Released 14 Apr 2025 · $0.10 in · $0.40 out
OpenAIGPT-5.4 nano Released 17 Mar 2026 · $0.20 in · $1.25 out
GPT-4.1 nano wins 0GPT-5.4 nano wins 17

Of the 17 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331).

GPT-4.1 nano or GPT-5.4 nano? The short answer

Pick GPT-5.4 nano if quality matters most: it scores higher (Intelligence Index 20.7 against 7.8). Pick GPT-4.1 nano if cost matters more: it costs 2.6 times less per token.

  • GPT-4.1 nano is cheaper: $0.18 against $0.46 per million tokens, 2.6 times less.
  • GPT-5.4 nano scores higher on the Artificial Analysis Intelligence Index: 20.7 against 7.8.
  • GPT-5.4 nano is newer: released 17 Mar 2026, against 14 Apr 2025 for GPT-4.1 nano.
  • On third-party benchmarks, on the results both models have: GPT-5.4 nano does better for agents and tool use, reasoning and knowledge, writing and sticking to the facts.

Head to head

7.8
Intelligence Index
20.7
$0.18
Price per million tokens, blendedlower is better
$0.46
1,047,576
Context window, tokens
400,000

Every fact side by side

GPT-4.1 nanoGPT-5.4 nano
Price per million tokens$0.10 in · $0.40 out$0.20 in · $1.25 out
Blended (3 in : 1 out)$0.18$0.46
Cheapest host, blended$0.18 (Azure)$0.46 (Azure)
Cached input, per million–$0.020 cached input
Intelligence Index7.820.7
Spring Prompt overall––
Output speed, tokens a second––
Context window1,047,576 tokens400,000 tokens
Released14 Apr 202517 Mar 2026 (newer)
WeightsClosedClosed
Developer based inthe USthe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 140 of 159 74 of 159
Reasoning and knowledgeHard questions across science, maths and general knowledge 159 of 187 96 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 143 of 169 103 of 169
Sticking to the factsSummarising without inventing things, and factual answers 137 of 139 82 of 139
Speed under pressureGood decisions against a real clock (fast chess) not measured 33 of 38

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureGPT-4.1 nanoGPT-5.4 nano
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 228 · 137 of 387 measured 1,322 1,401high reasoning
Business, management and finance · TextArena rating, higher is better 216 · 132 of 380 measured 1,321 1,405high reasoning
Creative writing · TextArena rating, higher is better 198 · 170 of 385 measured 1,307 1,336high reasoning
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 201 · 107 of 249 measured 7.8 20.7xhigh reasoning
GPQA Diamond% of questions, higher is better 201 · 103 of 232 measured 51.2% 81.7%xhigh reasoning
Humanity's Last Exam% of questions, higher is better 229 · 89 of 250 measured 3.8% 28.3%xhigh reasoning
SimpleQA Verified (Epoch AI) reported by SimpleQA Verified (Epoch AI) · 1 result
Correct answers% of questions, higher is better 76 · 72 of 76 measured 6.0% 11.7%high reasoning
Show all 17 results (10 more)
MeasureGPT-4.1 nanoGPT-5.4 nano
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Expert prompts · TextArena rating, higher is better 221 · 117 of 338 measured 1,313 1,437high reasoning
Instruction following · TextArena rating, higher is better 235 · 139 of 387 measured 1,300 1,387high reasoning
Writing, literature and language · TextArena rating, higher is better 231 · 156 of 386 measured 1,299 1,363high reasoning
Artificial Analysis reported by Artificial Analysis · 7 results
Terminal-Bench 2.1% of tasks, higher is better 119 · 50 of 131 measured 3.7% 60.7%xhigh reasoning
τ-bench banking% of tasks, higher is better 116 · 38 of 121 measured 3.5% 27.4%xhigh reasoning
Artificial Analysis Coding Indexindex score, higher is better 119 · 46 of 130 measured 11.1 56.1xhigh reasoning
IFBench% of instructions, higher is better 171 · 19 of 195 measured 32.0% 75.9%xhigh reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 209 · 78 of 241 measured 20.3% 76.7%xhigh reasoning
Terminal-Bench Hard% of tasks, higher is better 162 · 29 of 190 measured 3.8% 42.4%xhigh reasoning
τ²-bench telecom% of tasks, higher is better 173 · 75 of 193 measured 17.3% 76.0%xhigh reasoning

Quick answers

Is GPT-4.1 nano better than GPT-5.4 nano?

Pick GPT-5.4 nano if quality matters most: it scores higher (Intelligence Index 20.7 against 7.8). Pick GPT-4.1 nano if cost matters more: it costs 2.6 times less per token.

Which is cheaper, GPT-4.1 nano or GPT-5.4 nano?

GPT-4.1 nano is cheaper: $0.18 against $0.46 per million tokens, 2.6 times less.

Which is better for agents and tool use, GPT-4.1 nano or GPT-5.4 nano?

GPT-5.4 nano ranks 74 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 140 for GPT-4.1 nano.

Which is better for reasoning and knowledge, GPT-4.1 nano or GPT-5.4 nano?

GPT-5.4 nano ranks 96 of 187 models on sale for hard questions across science, maths and general knowledge, against 159 for GPT-4.1 nano.

Which is better for writing, GPT-4.1 nano or GPT-5.4 nano?

GPT-5.4 nano ranks 103 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 143 for GPT-4.1 nano.

Which is better for sticking to the facts, GPT-4.1 nano or GPT-5.4 nano?

GPT-5.4 nano ranks 82 of 139 models on sale for summarising without inventing things, and factual answers, against 137 for GPT-4.1 nano.

Which has the bigger context window, GPT-4.1 nano or GPT-5.4 nano?

GPT-4.1 nano: 1,047,576 tokens, against 400,000 for GPT-5.4 nano.

Run GPT-4.1 nano and GPT-5.4 nano on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.