Model comparison
GLM-5.3 vs Phi 3 Mini 4k Instruct
GLM-5.3 is the stronger model overall, scoring 54.8 to 27.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
MicrosoftSummary
- They share 18 benchmarks with published results for both. GLM-5.3 scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5.3 leads 75.7 to 27.6.
- The biggest single-benchmark swing is Chess Puzzles: 21% for GLM-5.3 and 0% for Phi 3 Mini 4k Instruct.
Side by side
| GLM-5.3 | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Microsoft |
| Noometry Index | 54.8 | 27.9 |
| Released | 2026-08-14 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 131K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $4.40 | — |
| Results tracked | 42 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.5 (#14), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1496 | 1093 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| CursorBench | 42.6% | — |
| LMArena WebDev | 1622 | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| WeirdML | 75.4% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 1,317 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
GLM-5.3: 36.4 (#38), Phi 3 Mini 4k Instruct: —
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| APEX-Agents | 56.6% | — |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 46.1 (#46), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 21% | 0% |
| LMArena Hard Prompts | 1489 | 1072 |
| NYT Connections (extended) | 74.2% | — |
| CritPt | 19.1% | — |
| LiveBench Reasoning | — | 26.8% |
| Mystery Game Puzzles | 33% | — |
| DTBench | 87.7% | — |
| LiveBench Data Analysis | — | 34.7% |
| LMCA | 55.5% | — |
| Adversarial NLI | — | 52.8% |
| Bench to the Future 3 | 0.15 | — |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 155.61 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math GLM-5.3 leads
GLM-5.3: 62.3 (#33), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1489 | 1111 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| OTIS Mock AIME 2024-2025 | 91.1% | — |
| ProofBench | 49% | — |
| LiveBench Math | — | 15.7% |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#37), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1516 | 1045 |
| GPQA Diamond | 90.9% | — |
| SimpleQA Verified | 41% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual GLM-5.3 leads
GLM-5.3: 55.7 (#28), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1457 | 1021 |
| LMArena Chinese | 1528 | 1021 |
| LMArena French | 1499 | 1076 |
| LMArena German | 1499 | 1044 |
| LMArena Japanese | 1453 | 935 |
| LMArena Korean | 1472 | 905 |
| LMArena Russian | 1463 | 1022 |
| LMArena Spanish | 1460 | 1085 |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.5 (#23), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1477 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context GLM-5.3 leads
GLM-5.3: 45.4 (#41), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1482 | 1044 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.7 (#6), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | GLM-5.3 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1471 | 1073 |
| LMArena Creative Writing | 1457 | 1037 |
| LMArena Multi-Turn | 1472 | 1018 |
| EQ-Bench Creative Writing | 2075 | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is GLM-5.3 better than Phi 3 Mini 4k Instruct?
GLM-5.3 is the stronger model overall, scoring 54.8 to 27.9 on the Noometry Index.
Is GLM-5.3 or Phi 3 Mini 4k Instruct better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.5 versus 26.6 in the Noometry coding category.
How many benchmarks do GLM-5.3 and Phi 3 Mini 4k Instruct share?
18 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.