Compare / Gemma 3 4B vs Ministral 3 3B

Gemma 3 4B vs Ministral 3 3B

Which is cheaper, which is more capable, and which does better at each kind of work, from 12 results on 2 sources that measured both models.

GoogleGemma 3 4B Released 11 Mar 2025 · $0.040 in · $0.050 out
Mistral AIMinistral 3 3B Released 2 Dec 2025 · $0.10 in · $0.10 out
Gemma 3 4B wins 41 too close to callMinistral 3 3B wins 7

Of the 12 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 Google vs Mistral AI, every model from both labs.

Gemma 3 4B or Ministral 3 3B? The short answer

They score about the same, so pick Gemma 3 4B: it costs 2.4 times less per token.

  • Gemma 3 4B is cheaper: $0.042 against $0.10 per million tokens, 2.4 times less.
  • They score about the same on the Artificial Analysis Intelligence Index (4.8 and 4.8).
  • Ministral 3 3B is newer: released 2 Dec 2025, against 11 Mar 2025 for Gemma 3 4B.
  • On third-party benchmarks, on the results both models have: Gemma 3 4B does better for sticking to the facts; Ministral 3 3B for reasoning and knowledge.

Head to head

4.8
Intelligence Index
4.8
$0.042
Price per million tokens, blendedlower is better
$0.10
131,072
Context window, tokens
131,072

Every fact side by side

Gemma 3 4BMinistral 3 3B
Price per million tokens$0.040 in · $0.050 out$0.10 in · $0.10 out
Blended (3 in : 1 out)$0.042$0.10
Cheapest host, blended$0.042 (Parasail, BF16)$0.10 (Mistral)
Cached input, per million–$0.010 cached input
Intelligence Index4.84.8
Spring Prompt overall––
Output speed, tokens a second–209
Context window131,072 tokens131,072 tokens
Released11 Mar 20252 Dec 2025 (newer)
WeightsOpenOpen
Developer based inthe USFrance

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 159 151 of 159
Reasoning and knowledgeHard questions across science, maths and general knowledge 181 of 187 174 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 157 of 169 not measured
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 4BMinistral 3 3B
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 246 · 246 of 249 measured 4.8 4.8
GPQA Diamond% of questions, higher is better 227 · 223 of 232 measured 29.1% 35.8%
Humanity's Last Exam% of questions, higher is better 194 · 193 of 250 measured 5.3% 5.4%
Vectara Hallucination Leaderboard reported by Vectara Hallucination Leaderboard · 2 results
Hallucination rate · Vectara% of summaries, lower is better 28 · 104 of 104 measured 6.4% 24.2%
Answer rate · Vectara% of documents, higher is better 103 · 102 of 104 measured 67.3% 74.3%
Show all 12 results (7 more)
MeasureGemma 3 4BMinistral 3 3B
Artificial Analysis reported by Artificial Analysis · 7 results
Terminal-Bench 2.1% of tasks, higher is better 127 · 130 of 131 measured 0.4% 0.0%
τ-bench banking% of tasks, higher is better 121 · 109 of 121 measured 0.4% 4.7%
Artificial Analysis Coding Indexindex score, higher is better 129 · 127 of 130 measured 2.7 4.8
IFBench% of instructions, higher is better 187 · 189 of 195 measured 28.3% 26.8%
Long-context reasoning (AA-LCR)% of questions, higher is better 226 · 216 of 241 measured 6.7% 17.0%
Terminal-Bench Hard% of tasks, higher is better 180 · 185 of 190 measured 0.8% 0.0%
τ²-bench telecom% of tasks, higher is better 187 · 153 of 193 measured 5.0% 24.9%

Quick answers

Is Gemma 3 4B better than Ministral 3 3B?

They score about the same, so pick Gemma 3 4B: it costs 2.4 times less per token.

Which is cheaper, Gemma 3 4B or Ministral 3 3B?

Gemma 3 4B is cheaper: $0.042 against $0.10 per million tokens, 2.4 times less.

Which is better for agents and tool use, Gemma 3 4B or Ministral 3 3B?

Ministral 3 3B ranks 151 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 Ministral 3 3B?

Ministral 3 3B ranks 174 of 187 models on sale for hard questions across science, maths and general knowledge, against 181 for Gemma 3 4B.

Which is better for sticking to the facts, Gemma 3 4B or Ministral 3 3B?

Gemma 3 4B ranks 32 of 139 models on sale for summarising without inventing things, and factual answers, against 137 for Ministral 3 3B.

Run Gemma 3 4B and Ministral 3 3B 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.