Compare / Phi 4 vs Ministral 3 14B 2512
Phi 4 vs Ministral 3 14B 2512
Which is cheaper, which is more capable, and which does better at each kind of work, from 14 results on 3 sources that measured both models.
Phi 4 or Ministral 3 14B 2512? The short answer
- Phi 4 is cheaper: $0.088 against $0.20 per million tokens, 2.3 times less.
- They score about the same on the Artificial Analysis Intelligence Index (5.9 and 6.0).
- Ministral 3 14B 2512 writes faster: 91 tokens a second against 42.
- On our business benchmarks, Phi 4 ranks higher for agents and tool use and sticking to the facts.
- On our business benchmarks, Ministral 3 14B 2512 ranks higher for reasoning and knowledge and writing.
At a glance
| Phi 4 | Ministral 3 14B 2512 | |
|---|---|---|
| Price per million tokens | $0.070 in · $0.14 out | $0.20 in · $0.20 out |
| Blended (3 in : 1 out) | $0.088 | $0.20 |
| Cheapest host, blended | – | $0.20 (Mistral) |
| Intelligence Index | 5.9 | 6.0 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | 42 | 91 |
| Context window | 16,384 tokens | 262,144 tokens |
| Weights | Open | Open |
| Developer based in | the US | France |
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 models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.
| Use case | Phi 4 | Ministral 3 14B 2512 |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 122 of 161 | 128 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 178 of 195 | 167 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 175 of 187 | 143 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | 3 of 153 | 146 of 153 |
Every shared result
| Benchmark | Metric | Phi 4 | Ministral 3 14B 2512 |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 5.90 | 6.00 |
| Artificial Analysisreported | GPQA Diamond% of questions | 57.5% | 57.2% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 3.8% | 4.6% |
| Artificial Analysisreported | IFBench% of instructions | 23.5% | 32.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 0.0% | 26.3% |
| Artificial Analysisreported | Output speedtokens per second | 42.3 | 91.3 |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 3.8% | 4.5% |
| Artificial Analysisreported | Time to first answer tokenseconds | 1.03 | 0.50 |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 0.0% | 27.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 19.0% | 24.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.34 | 0.32 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 25.7 | 31.4 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 80.7% | 99.6% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 3.7% | 19.4% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.
Quick answers
Is Phi 4 better than Ministral 3 14B 2512?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Phi 4 ranks higher for agents and tool use and sticking to the facts; Ministral 3 14B 2512 ranks higher for reasoning and knowledge and writing.
Which is cheaper, Phi 4 or Ministral 3 14B 2512?
Phi 4 costs $0.088 per million tokens (three input to one output), against $0.20 for Ministral 3 14B 2512.
Which is faster, Phi 4 or Ministral 3 14B 2512?
Ministral 3 14B 2512 writes about 91 tokens a second on its usual API, against 42 for Phi 4 (Artificial Analysis).
Run Phi 4 and Ministral 3 14B 2512 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.