Compare / Nova Pro 1.0 vs Gemma 3 4B

Nova Pro 1.0 vs Gemma 3 4B

Which is cheaper, which is more capable, and which does better at each kind of work, from 23 results on 4 sources that measured both models.

AmazonNova Pro 1.0 Released 3 Dec 2024 · $0.80 in · $3.20 out
GoogleGemma 3 4B Released 11 Mar 2025 · $0.040 in · $0.050 out
Nova Pro 1.0 wins 144 too close to callGemma 3 4B wins 5

Of the 23 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also Amazon vs Google, every model from both labs.

Nova Pro 1.0 or Gemma 3 4B? The short answer

Pick Nova Pro 1.0 if quality matters most: it scores higher (Intelligence Index 7.0 against 4.8). Pick Gemma 3 4B if cost matters more: it costs 33 times less per token.

  • Gemma 3 4B is cheaper: $0.042 against $1.40 per million tokens, 33 times less.
  • Nova Pro 1.0 scores higher on the Artificial Analysis Intelligence Index: 7.0 against 4.8.
  • Gemma 3 4B is newer: released 11 Mar 2025, against 3 Dec 2024 for Nova Pro 1.0.
  • On third-party benchmarks, on the results both models have: Nova Pro 1.0 does better for agents and tool use, reasoning and knowledge and sticking to the facts.

Head to head

7.0
Intelligence Index
4.8
$1.40
Price per million tokens, blendedlower is better
$0.042
300,000
Context window, tokens
131,072

Every fact side by side

Nova Pro 1.0Gemma 3 4B
Price per million tokens$0.80 in · $3.20 out$0.040 in · $0.050 out
Blended (3 in : 1 out)$1.40$0.042
Cheapest host, blended$1.40 (Amazon Bedrock)$0.042 (Parasail, BF16)
Intelligence Index7.04.8
Spring Prompt overall––
Output speed, tokens a second––
Context window300,000 tokens131,072 tokens
Released3 Dec 202411 Mar 2025 (newer)
WeightsClosedOpen
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 149 of 159 158 of 159
Reasoning and knowledgeHard questions across science, maths and general knowledge 173 of 187 181 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 156 of 169 157 of 169
Sticking to the factsSummarising without inventing things, and factual answers 8 of 139 32 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.

MeasureNova Pro 1.0Gemma 3 4B
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 261 · 255 of 387 measured 1,290 1,303
Business, management and finance · TextArena rating, higher is better 267 · 208 of 380 measured 1,279 1,330
Creative writing · TextArena rating, higher is better 278 · 243 of 385 measured 1,237 1,275
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 218 · 246 of 249 measured 7.0 4.8
GPQA Diamond% of questions, higher is better 203 · 227 of 232 measured 49.9% 29.1%
Humanity's Last Exam% of questions, higher is better 244 · 194 of 250 measured 3.2% 5.3%
Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 3 results
Overall accuracy · BFCL% correct, higher is better 65 · 74 of 81 measured 25.0%bfcl fc 19.6%bfcl prompt
Irrelevance detection · BFCL% correct, higher is better 63 · 71 of 81 measured 70.1%bfcl fc 53.9%bfcl prompt
Memory · BFCL% correct, higher is better 75 · 53 of 81 measured 1.9%bfcl fc 8.6%bfcl prompt
Vectara Hallucination Leaderboard reported by Vectara Hallucination Leaderboard · 2 results
Hallucination rate · Vectara% of summaries, lower is better 10 · 28 of 104 measured 5.1% 6.4%
Answer rate · Vectara% of documents, higher is better 58 · 103 of 104 measured 99.3% 67.3%
Show all 23 results (12 more)
MeasureNova Pro 1.0Gemma 3 4B
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Expert prompts · TextArena rating, higher is better 258 · 270 of 338 measured 1,273 1,260
Instruction following · TextArena rating, higher is better 263 · 270 of 387 measured 1,277 1,267
Writing, literature and language · TextArena rating, higher is better 263 · 237 of 386 measured 1,274 1,298
Artificial Analysis reported by Artificial Analysis · 4 results
IFBench% of instructions, higher is better 144 · 187 of 195 measured 38.1% 28.3%
Long-context reasoning (AA-LCR)% of questions, higher is better 207 · 226 of 241 measured 21.0% 6.7%
Terminal-Bench Hard% of tasks, higher is better 147 · 180 of 190 measured 6.1% 0.8%
τ²-bench telecom% of tasks, higher is better 180 · 187 of 193 measured 14.0% 5.0%
Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 5 results
Multi-turn tasks · BFCL% correct, higher is better 72 · 75 of 81 measured 1.9%bfcl fc 0.4%bfcl prompt
Relevance detection · BFCL% correct, higher is better 8 · 1 of 81 measured 93.8%bfcl fc 100.0%bfcl prompt
Single-turn calls (curated) · BFCL% correct, higher is better 30 · 74 of 81 measured 86.6%bfcl fc 61.1%bfcl prompt
Single-turn calls (user-contributed) · BFCL% correct, higher is better 19 · 64 of 81 measured 78.5%bfcl fc 60.8%bfcl prompt
Web search · BFCL% correct, higher is better 48 · 56 of 81 measured 2.5%bfcl fc 1.0%bfcl prompt

Quick answers

Is Nova Pro 1.0 better than Gemma 3 4B?

Pick Nova Pro 1.0 if quality matters most: it scores higher (Intelligence Index 7.0 against 4.8). Pick Gemma 3 4B if cost matters more: it costs 33 times less per token.

Which is cheaper, Nova Pro 1.0 or Gemma 3 4B?

Gemma 3 4B is cheaper: $0.042 against $1.40 per million tokens, 33 times less.

Which is better for agents and tool use, Nova Pro 1.0 or Gemma 3 4B?

Nova Pro 1.0 ranks 149 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, Nova Pro 1.0 or Gemma 3 4B?

Nova Pro 1.0 ranks 173 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, Nova Pro 1.0 or Gemma 3 4B?

Nova Pro 1.0 ranks 156 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, Nova Pro 1.0 or Gemma 3 4B?

Nova Pro 1.0 ranks 8 of 139 models on sale for summarising without inventing things, and factual answers, against 32 for Gemma 3 4B.

Which has the bigger context window, Nova Pro 1.0 or Gemma 3 4B?

Nova Pro 1.0: 300,000 tokens, against 131,072 for Gemma 3 4B.

Run Nova Pro 1.0 and Gemma 3 4B 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.