Compare / Gemma 3 4B vs Devstral Small 2

Gemma 3 4B vs Devstral Small 2

Which is cheaper, which is more capable, and which does better at each kind of work, from 13 results on 2 sources that measured both models.

GoogleGemma 3 4B Released 11 Mar 2025 · $0.040 in · $0.050 out
Mistral AIDevstral Small 2 Released 9 Dec 2025 · Not on sale
Gemma 3 4B wins 2Devstral Small 2 wins 11

Of the 13 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 Devstral Small 2? The short answer

Pick Devstral Small 2: it scores higher (Intelligence Index 7.5 against 4.8) at about the same price.

  • Devstral Small 2 scores higher on the Artificial Analysis Intelligence Index: 7.5 against 4.8.
  • Devstral Small 2 is newer: released 9 Dec 2025, against 11 Mar 2025 for Gemma 3 4B.
  • On third-party benchmarks, on the results both models have: Devstral Small 2 does better for agents and tool use, reasoning and knowledge and writing.

Head to head

4.8
Intelligence Index
7.5

Every fact side by side

Gemma 3 4BDevstral Small 2
Price per million tokens$0.040 in · $0.050 outNot on sale
Blended (3 in : 1 out)$0.042–
Cheapest host, blended$0.042 (Parasail, BF16)–
Intelligence Index4.87.5
Spring Prompt overall––
Output speed, tokens a second––
Context window131,072 tokens–
Released11 Mar 20259 Dec 2025 (newer)
WeightsOpenClosed
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 159 114 of 160
Reasoning and knowledgeHard questions across science, maths and general knowledge 181 of 187 162 of 188
WritingWhat people prefer in blind comparisons, and judged writing quality 157 of 169 133 of 170
Sticking to the factsSummarising without inventing things, and factual answers 32 of 139 not measured

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 4BDevstral Small 2
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 246 · 211 of 249 measured 4.8 7.5
GPQA Diamond% of questions, higher is better 227 · 194 of 232 measured 29.1% 53.2%
Humanity's Last Exam% of questions, higher is better 194 · 242 of 250 measured 5.3% 3.5%
UGI Leaderboard reported by UGI Leaderboard · 3 results
Writing score · UGIscore out of 100, higher is better 185 · 146 of 230 measured 24.0 31.6
Requested-length error · UGI% off the requested word count, lower is better 33 · 151 of 230 measured 9.0% 26.0%
Style adherence · UGIscore from 0 to 1, higher is better 214 · 113 of 230 measured 0.29 0.35
Show all 13 results (7 more)
MeasureGemma 3 4BDevstral Small 2
Artificial Analysis reported by Artificial Analysis · 7 results
Terminal-Bench 2.1% of tasks, higher is better 127 · 83 of 131 measured 0.4% 29.6%
τ-bench banking% of tasks, higher is better 121 · 74 of 121 measured 0.4% 10.7%
Artificial Analysis Coding Indexindex score, higher is better 129 · 86 of 130 measured 2.7 29.3
IFBench% of instructions, higher is better 187 · 177 of 195 measured 28.3% 31.2%
Long-context reasoning (AA-LCR)% of questions, higher is better 226 · 196 of 241 measured 6.7% 28.0%
Terminal-Bench Hard% of tasks, higher is better 180 · 105 of 190 measured 0.8% 16.7%
τ²-bench telecom% of tasks, higher is better 187 · 157 of 193 measured 5.0% 23.4%

Quick answers

Is Gemma 3 4B better than Devstral Small 2?

Pick Devstral Small 2: it scores higher (Intelligence Index 7.5 against 4.8) at about the same price.

Which is better for agents and tool use, Gemma 3 4B or Devstral Small 2?

Devstral Small 2 ranks 114 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 Devstral Small 2?

Devstral Small 2 ranks 162 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 Devstral Small 2?

Devstral Small 2 ranks 133 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 157 for Gemma 3 4B.

Run Gemma 3 4B and Devstral Small 2 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.