# Granite 3.1 8b Instruct vs Kimi K3

> Kimi K3 is the stronger model overall, scoring 59.5 to 32.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/granite-3-1-8b-instruct-vs-kimi-k3
- Last updated: 2026-10-10
- Shared benchmarks: 12

## Summary

- They share 12 benchmarks with published results for both. Granite 3.1 8b Instruct scores higher in 0 categories and Kimi K3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 33.0.

## Snapshot

| | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| Provider | IBM | Moonshot AI |
| Noometry Index | 32.4 | 59.5 |
| Rank | 258 | 15 |
| Context | — | 1.05M |
| Input $/M | — | $3 |
| Output $/M | — | $15 |
| Weights | Open | Open |

## Coding

- Granite 3.1 8b Instruct: 34.5 (#233)
- Kimi K3: 61.0 (#10)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Coding | 1186 | 1508 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| WeirdML | — | 82.6% |
| ALE-Bench | — | 1,524 |

## Agentic & Tool Use

- Granite 3.1 8b Instruct: 24.1 (#120)
- Kimi K3: 41.8 (#20)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| Berkeley Function Calling Leaderboard | 27.1% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |

## Reasoning

- Granite 3.1 8b Instruct: 22.1 (#207)
- Kimi K3: 63.0 (#17)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Hard Prompts | 1145 | 1496 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Chess Puzzles | — | 39% |
| Mystery Game Puzzles | — | 26% |
| DTBench | — | 91.2% |
| LMCA | — | 52.7% |
| Surface Evolver Bench | — | 95% |
| Epoch Capabilities Index | — | 157.45 |
| ForecastBench | — | 61.1 |

## Math

- Granite 3.1 8b Instruct: 33.0 (#209)
- Kimi K3: 74.2 (#16)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Math | 1152 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| OTIS Mock AIME 2024-2025 | — | 97.2% |
| ProofBench | — | 87% |

## Knowledge

- Granite 3.1 8b Instruct: 31.1 (#220)
- Kimi K3: 63.2 (#21)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Expert | 1142 | 1521 |
| GPQA Diamond | — | 93.1% |
| SimpleQA Verified | — | 50.6% |

## Multimodal

- Granite 3.1 8b Instruct: —
- Kimi K3: 37.8 (#70)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |

## Multilingual

- Granite 3.1 8b Instruct: 30.9 (#260)
- Kimi K3: 56.3 (#21)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1099 | 1466 |
| LMArena Chinese | 1145 | 1529 |
| LMArena Russian | 1092 | 1482 |
| LMArena French | — | 1491 |
| LMArena German | — | 1488 |
| LMArena Japanese | — | 1487 |
| LMArena Korean | — | 1458 |
| LMArena Spanish | — | 1472 |

## Instruction Following

- Granite 3.1 8b Instruct: 58.6 (#259)
- Kimi K3: 77.7 (#14)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1131 | 1483 |

## Long Context

- Granite 3.1 8b Instruct: 35.2 (#241)
- Kimi K3: 45.8 (#29)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1162 | 1494 |

## Writing & Preference

- Granite 3.1 8b Instruct: 35.5 (#266)
- Kimi K3: 76.6 (#4)

| Benchmark | Granite 3.1 8b Instruct | Kimi K3 |
|---|---|---|
| LMArena Text | 1150 | 1476 |
| LMArena Creative Writing | 1129 | 1454 |
| LMArena Multi-Turn | 1108 | 1488 |
| EQ-Bench Creative Writing | — | 2082 |
| EQ-Bench 4 | — | 1339 |

## FAQ

### Is Granite 3.1 8b Instruct better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 32.4 on the Noometry Index.

### Is Granite 3.1 8b Instruct or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 34.5 in the Noometry coding category.

### How many benchmarks do Granite 3.1 8b Instruct and Kimi K3 share?

12 benchmarks have published results for both models. Granite 3.1 8b Instruct has 13 scored results on Noometry and Kimi K3 has 53.