Compare / Ministral 3 14B vs gpt-oss-20b
Ministral 3 14B vs gpt-oss-20b
Which is cheaper, which is more capable, and which does better at each kind of work, from 17 results on 2 sources that measured both models.
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 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also Mistral AI vs OpenAI, every model from both labs.
Ministral 3 14B or gpt-oss-20b? The short answer
Pick gpt-oss-20b: it scores higher (Intelligence Index 10.0 against 6.0) and costs less.
- gpt-oss-20b is cheaper: $0.058 against $0.20 per million tokens, 3.5 times less.
- gpt-oss-20b scores higher on the Artificial Analysis Intelligence Index: 10.0 against 6.0.
- gpt-oss-20b writes faster: 277 tokens a second against 79.
- Ministral 3 14B is newer: released 2 Dec 2025, against 5 Aug 2025 for gpt-oss-20b.
- On third-party benchmarks, on the results both models have: Ministral 3 14B does better for writing; gpt-oss-20b for agents and tool use and reasoning and knowledge.
Head to head
Every fact side by side
| Ministral 3 14B | gpt-oss-20b | |
|---|---|---|
| Price per million tokens | $0.20 in · $0.20 out | $0.030 in · $0.14 out |
| Blended (3 in : 1 out) | $0.20 | $0.058 |
| Cheapest host, blended | $0.20 (Mistral) | $0.022 (PrimeIntellect, FP4) |
| Cached input, per million | $0.020 cached input | – |
| Intelligence Index | 6.0 | 10.0 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | 79 | 277 |
| Context window | 262,144 tokens | 131,072 tokens |
| Released | 2 Dec 2025 (newer) | 5 Aug 2025 |
| Weights | Open | Open |
| Developer based in | France | the 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.
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.
| Measure | Ministral 3 14B | gpt-oss-20b |
|---|---|---|
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 234 · 176 of 249 measured | 6.0 | 10.0low reasoning |
| GPQA Diamond% of questions, higher is better 192 · 159 of 232 measured | 57.2% | 68.8%high reasoning |
| Humanity's Last Exam% of questions, higher is better 206 · 153 of 250 measured | 4.6% | 11.0%high reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 148 · 182 of 230 measured | 31.4 | 24.6low reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 140 · 221 of 230 measured | 24.0% | 98.0%low reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 167 · 187 of 230 measured | 0.32 | 0.31medium reasoning |
Show all 17 results (11 more)
| Measure | Ministral 3 14B | gpt-oss-20b |
|---|---|---|
| Artificial Analysis reported by Artificial Analysis · 11 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 106 · 97 of 131 measured | 9.7% | 13.9%high reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 72 · 72 of 120 measured | 0.0% | 0.0%high reasoning |
| τ-bench banking% of tasks, higher is better 93 · 90 of 121 measured | 6.6% | 7.0%high reasoning |
| Artificial Analysis Coding Indexindex score, higher is better 112 · 100 of 130 measured | 14.4 | 20.7high reasoning |
| IFBench% of instructions, higher is better 171 · 69 of 195 measured | 32.0% | 65.1%high reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 199 · 184 of 241 measured | 26.3% | 34.7%high reasoning |
| Output speedtokens per second, higher is better 53 · 6 of 78 measured | 79.2 | 277low reasoning |
| SciCode% of problems, higher is better 116 · 90 of 121 measured | 23.8% | 38.9%high reasoning |
| Terminal-Bench Hard% of tasks, higher is better 156 · 127 of 190 measured | 4.5% | 10.6%high reasoning |
| Time to first answer tokenseconds, lower is better 7 · 44 of 78 measured | 0.52 | 7.62low reasoning |
| τ²-bench telecom% of tasks, higher is better 144 · 96 of 193 measured | 27.2% | 60.2%high reasoning |
Compare them with other models
Quick answers
Is Ministral 3 14B better than gpt-oss-20b?
Pick gpt-oss-20b: it scores higher (Intelligence Index 10.0 against 6.0) and costs less.
Which is cheaper, Ministral 3 14B or gpt-oss-20b?
gpt-oss-20b is cheaper: $0.058 against $0.20 per million tokens, 3.5 times less.
Which is faster, Ministral 3 14B or gpt-oss-20b?
gpt-oss-20b writes about 277 tokens a second on its usual API, against 79 for Ministral 3 14B (Artificial Analysis).
Which is better for agents and tool use, Ministral 3 14B or gpt-oss-20b?
gpt-oss-20b ranks 130 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 134 for Ministral 3 14B.
Which is better for reasoning and knowledge, Ministral 3 14B or gpt-oss-20b?
gpt-oss-20b ranks 132 of 187 models on sale for hard questions across science, maths and general knowledge, against 157 for Ministral 3 14B.
Which is better for writing, Ministral 3 14B or gpt-oss-20b?
Ministral 3 14B ranks 135 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 155 for gpt-oss-20b.
Which has the bigger context window, Ministral 3 14B or gpt-oss-20b?
Ministral 3 14B: 262,144 tokens, against 131,072 for gpt-oss-20b.
Run Ministral 3 14B and gpt-oss-20b 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.