Model comparison
Granite 4.0 Micro vs Qwen3 14B
Qwen3 14B is the stronger model overall, scoring 35.5 to 29.0 on the Noometry Index. Granite 4.0 Micro costs 15× less per token, which makes it the better buy when Qwen3 14B's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Granite 4.0 Micro scores higher in 1 category and Qwen3 14B in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 14B leads 39.3 to 9.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.8% for Granite 4.0 Micro and 66.4% for Qwen3 14B.
- Granite 4.0 Micro is cheaper at $0.017 / $0.11 per million input/output tokens, against $0.35 / $1.40 for Qwen3 14B.
- Qwen3 14B accepts more context: 131K tokens versus 131K.
Side by side
| Granite 4.0 Micro | Qwen3 14B | |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 29.0 | 35.5 |
| Released | 2025-10-02 | 2025-04 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 118K | 8K |
| Input $ / M tokens | $0.017 | $0.35 |
| Output $ / M tokens | $0.11 | $1.40 |
| Results tracked | 8 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Granite 4.0 Micro: —, Qwen3 14B: 37.3 (#195)
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| SciCode | — | 31.6% |
Agentic & Tool Use Not comparable
Granite 4.0 Micro: —, Qwen3 14B: 29.6 (#83)
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41% |
Reasoning Too close to call
Granite 4.0 Micro: 19.2 (#265), Qwen3 14B: 18.5 (#280)
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| Chess Puzzles | 0% | 4% |
| Kagi LLM Benchmark | — | 49.1% |
| CritPt | — | 0% |
| DTBench | — | 64% |
| LMCA | — | 18.2% |
| Epoch Capabilities Index | — | 138.23 |
Math Qwen3 14B leads
Granite 4.0 Micro: 12.0 (#307), Qwen3 14B: 38.6 (#133)
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.8% | 66.4% |
| Omni-MATH | 20.9% | — |
Knowledge Qwen3 14B leads
Granite 4.0 Micro: 9.9 (#304), Qwen3 14B: 39.3 (#134)
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| GPQA Diamond | 28.3% | 63.8% |
| MMLU-Pro | 39.5% | — |
| Vectara Hallucination Rate | — | 5.4% |
| GPQA (HELM) | 30.7% | — |
Instruction Following Not comparable
Granite 4.0 Micro: 69.9 (#169), Qwen3 14B: —
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| IFEval | 84.9% | — |
Long Context Not comparable
Granite 4.0 Micro: —, Qwen3 14B: 38.1 (#204)
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
Writing & Preference Not comparable
Granite 4.0 Micro: 46.7 (#216), Qwen3 14B: —
| Benchmark | Granite 4.0 Micro | Qwen3 14B |
|---|---|---|
| WildBench | 67% | — |
Frequently asked questions
Is Granite 4.0 Micro better than Qwen3 14B?
Qwen3 14B is the stronger model overall, scoring 35.5 to 29.0 on the Noometry Index. Granite 4.0 Micro costs 15× less per token, which makes it the better buy when Qwen3 14B's lead doesn't matter for your workload.
Which is cheaper, Granite 4.0 Micro or Qwen3 14B?
Granite 4.0 Micro is cheaper. It lists at $0.017 per million input tokens and $0.11 per million output tokens; Qwen3 14B lists at $0.35 and $1.40.
Which has the bigger context window?
Qwen3 14B does, with 131K tokens against 131K.
How many benchmarks do Granite 4.0 Micro and Qwen3 14B share?
3 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and Qwen3 14B has 12.