Compare / Gemma 3 4B vs GPT-4.1 nano

Gemma 3 4B vs GPT-4.1 nano

Which is cheaper, which is more capable, and which does better at each kind of work, from 24 results on 3 sources that measured both models.

GoogleGemma 3 4B Released 11 Mar 2025 · $0.040 in · $0.050 out
OpenAIGPT-4.1 nano Released 14 Apr 2025 · $0.10 in · $0.40 out
Gemma 3 4B wins 34 too close to callGPT-4.1 nano wins 17

Of the 24 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). See also Google vs OpenAI, every model from both labs.

Gemma 3 4B or GPT-4.1 nano? The short answer

Pick GPT-4.1 nano if quality matters most: it scores higher (Intelligence Index 7.8 against 4.8). Pick Gemma 3 4B if cost matters more: it costs 4.1 times less per token.

  • Gemma 3 4B is cheaper: $0.042 against $0.18 per million tokens, 4.1 times less.
  • GPT-4.1 nano scores higher on the Artificial Analysis Intelligence Index: 7.8 against 4.8.
  • GPT-4.1 nano is newer: released 14 Apr 2025, against 11 Mar 2025 for Gemma 3 4B.
  • On third-party benchmarks, on the results both models have: GPT-4.1 nano does better for agents and tool use, reasoning and knowledge and writing.

Head to head

4.8
Intelligence Index
7.8
$0.042
Price per million tokens, blendedlower is better
$0.18
131,072
Context window, tokens
1,047,576

Every fact side by side

Gemma 3 4BGPT-4.1 nano
Price per million tokens$0.040 in · $0.050 out$0.10 in · $0.40 out
Blended (3 in : 1 out)$0.042$0.18
Cheapest host, blended$0.042 (Parasail, BF16)$0.18 (Azure)
Intelligence Index4.87.8
Spring Prompt overall––
Output speed, tokens a second––
Context window131,072 tokens1,047,576 tokens
Released11 Mar 202514 Apr 2025 (newer)
WeightsOpenClosed
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 158 of 159 140 of 159
Reasoning and knowledgeHard questions across science, maths and general knowledge 181 of 187 159 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 157 of 169 143 of 169
Sticking to the factsSummarising without inventing things, and factual answers 32 of 139 137 of 139

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.

MeasureGemma 3 4BGPT-4.1 nano
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 255 · 228 of 387 measured 1,303 1,322
Business, management and finance · TextArena rating, higher is better 208 · 216 of 380 measured 1,330 1,321
Creative writing · TextArena rating, higher is better 243 · 198 of 385 measured 1,275 1,307
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 246 · 201 of 249 measured 4.8 7.8
GPQA Diamond% of questions, higher is better 227 · 201 of 232 measured 29.1% 51.2%
Humanity's Last Exam% of questions, higher is better 194 · 229 of 250 measured 5.3% 3.8%
Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 3 results
Overall accuracy · BFCL% correct, higher is better 74 · 42 of 81 measured 19.6%bfcl prompt 33.1%bfcl fc
Irrelevance detection · BFCL% correct, higher is better 71 · 33 of 81 measured 53.9%bfcl prompt 83.4%bfcl prompt
Memory · BFCL% correct, higher is better 53 · 34 of 81 measured 8.6%bfcl prompt 18.9%bfcl fc
Show all 24 results (15 more)
MeasureGemma 3 4BGPT-4.1 nano
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Expert prompts · TextArena rating, higher is better 270 · 221 of 338 measured 1,260 1,313
Instruction following · TextArena rating, higher is better 270 · 235 of 387 measured 1,267 1,300
Writing, literature and language · TextArena rating, higher is better 237 · 231 of 386 measured 1,298 1,299
Artificial Analysis reported by Artificial Analysis · 7 results
Terminal-Bench 2.1% of tasks, higher is better 127 · 119 of 131 measured 0.4% 3.7%
τ-bench banking% of tasks, higher is better 121 · 116 of 121 measured 0.4% 3.5%
Artificial Analysis Coding Indexindex score, higher is better 129 · 119 of 130 measured 2.7 11.1
IFBench% of instructions, higher is better 187 · 171 of 195 measured 28.3% 32.0%
Long-context reasoning (AA-LCR)% of questions, higher is better 226 · 209 of 241 measured 6.7% 20.3%
Terminal-Bench Hard% of tasks, higher is better 180 · 162 of 190 measured 0.8% 3.8%
τ²-bench telecom% of tasks, higher is better 187 · 173 of 193 measured 5.0% 17.3%
Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 5 results
Multi-turn tasks · BFCL% correct, higher is better 75 · 39 of 81 measured 0.4%bfcl prompt 23.6%bfcl fc
Relevance detection · BFCL% correct, higher is better 1 · 8 of 81 measured 100.0%bfcl prompt 93.8%bfcl fc
Single-turn calls (curated) · BFCL% correct, higher is better 74 · 66 of 81 measured 61.1%bfcl prompt 73.0%bfcl fc
Single-turn calls (user-contributed) · BFCL% correct, higher is better 64 · 65 of 81 measured 60.8%bfcl prompt 60.8%bfcl fc
Web search · BFCL% correct, higher is better 56 · 36 of 81 measured 1.0%bfcl prompt 11.0%bfcl fc

Quick answers

Is Gemma 3 4B better than GPT-4.1 nano?

Pick GPT-4.1 nano if quality matters most: it scores higher (Intelligence Index 7.8 against 4.8). Pick Gemma 3 4B if cost matters more: it costs 4.1 times less per token.

Which is cheaper, Gemma 3 4B or GPT-4.1 nano?

Gemma 3 4B is cheaper: $0.042 against $0.18 per million tokens, 4.1 times less.

Which is better for agents and tool use, Gemma 3 4B or GPT-4.1 nano?

GPT-4.1 nano ranks 140 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 158 for Gemma 3 4B.

Which is better for reasoning and knowledge, Gemma 3 4B or GPT-4.1 nano?

GPT-4.1 nano ranks 159 of 187 models on sale for hard questions across science, maths and general knowledge, against 181 for Gemma 3 4B.

Which is better for writing, Gemma 3 4B or GPT-4.1 nano?

GPT-4.1 nano ranks 143 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 157 for Gemma 3 4B.

Which is better for sticking to the facts, Gemma 3 4B or GPT-4.1 nano?

Gemma 3 4B ranks 32 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, Gemma 3 4B or GPT-4.1 nano?

GPT-4.1 nano: 1,047,576 tokens, against 131,072 for Gemma 3 4B.

Run Gemma 3 4B and GPT-4.1 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.