Model comparison
MiniMax-M2.5 vs Qwen3 14B
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 35.0 on the Noometry Index.
Last verified . 3 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. MiniMax-M2.5 scores higher in 2 categories and Qwen3 14B in 4 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 47.9 to 36.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 55.2% for MiniMax-M2.5 and 49.1% for Qwen3 14B.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.35 / $1.40 for Qwen3 14B.
- MiniMax-M2.5 accepts more context: 205K tokens versus 131K.
Side by side
| MiniMax-M2.5 | Qwen3 14B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 38.1 | 35.0 |
| Released | 2026-02-12 | 2025-04 |
| Weights | Open | Open |
| Context window | 205K | 131K |
| Max output | 131K | 8K |
| Input $ / M tokens | $0.30 | $0.35 |
| Output $ / M tokens | $1.20 | $1.40 |
| Results tracked | 33 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
MiniMax-M2.5: 47.9 (#53), Qwen3 14B: 36.7 (#191)
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1386 | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 31.6% |
| LMArena Coding | 1380 | — |
| ALE-Bench | 618.17 | — |
Agentic & Tool Use Too close to call
MiniMax-M2.5: 30.2 (#70), Qwen3 14B: 29.4 (#78)
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Berkeley Function Calling Leaderboard | — | 41% |
| Vending-Bench 2 | -23.16 | — |
Reasoning Too close to call
MiniMax-M2.5: 17.6 (#281), Qwen3 14B: 18.1 (#272)
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| Kagi LLM Benchmark | 55.2% | 49.1% |
| Epoch Capabilities Index | 146.68 | 138.23 |
| ARC-AGI-2 | 4.9% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 0% |
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1372 | — |
| DTBench | — | 64% |
| LMCA | — | 18.2% |
Math Qwen3 14B leads
MiniMax-M2.5: 25.8 (#242), Qwen3 14B: 37.0 (#102)
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.4% |
| ProofBench | 4% | — |
| LMArena Math | 1378 | — |
Knowledge Too close to call
MiniMax-M2.5: 38.6 (#132), Qwen3 14B: 39.2 (#124)
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| Vectara Hallucination Rate | 9.1% | 5.4% |
| GPQA Diamond | — | 63.8% |
| LMArena Expert | 1378 | — |
Multilingual Not comparable
MiniMax-M2.5: 47.3 (#147), Qwen3 14B: —
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| LMArena Non-English | 1338 | — |
| LMArena Chinese | 1392 | — |
| LMArena French | 1363 | — |
| LMArena German | 1361 | — |
| LMArena Japanese | 1169 | — |
| LMArena Korean | 1232 | — |
| LMArena Russian | 1359 | — |
| LMArena Spanish | 1354 | — |
Instruction Following Not comparable
MiniMax-M2.5: 71.7 (#143), Qwen3 14B: —
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| LMArena Instruction Following | 1353 | — |
Long Context Too close to call
MiniMax-M2.5: 37.4 (#211), Qwen3 14B: 38.0 (#201)
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
| LMArena Longer Query | 1367 | — |
Writing & Preference Not comparable
MiniMax-M2.5: 54.1 (#149), Qwen3 14B: —
| Benchmark | MiniMax-M2.5 | Qwen3 14B |
|---|---|---|
| LMArena Text | 1359 | — |
| LMArena Creative Writing | 1331 | — |
| EQ-Bench Creative Writing | 1361 | — |
| LMArena Multi-Turn | 1364 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Qwen3 14B?
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 35.0 on the Noometry Index.
Which is cheaper, MiniMax-M2.5 or Qwen3 14B?
MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3 14B lists at $0.35 and $1.40.
Is MiniMax-M2.5 or Qwen3 14B better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 36.7 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 131K.
How many benchmarks do MiniMax-M2.5 and Qwen3 14B share?
3 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Qwen3 14B has 12.