Model comparison
Granite 3.1 8b Instruct vs Mercury 2
Mercury 2 is the stronger model overall, scoring 38.8 to 31.8 on the Noometry Index.
Last verified . 11 shared benchmarks.
Summary
- They share 11 benchmarks with published results for both. Granite 3.1 8b Instruct scores higher in 1 category and Mercury 2 in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mercury 2 leads 54.0 to 35.7.
- Granite 3.1 8b Instruct has downloadable open weights; the other is API-only.
Side by side
| Granite 3.1 8b Instruct | Mercury 2 | |
|---|---|---|
| Provider | IBM | Inception |
| Noometry Index | 31.8 | 38.8 |
| Released | — | 2026-02-20 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 50K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.75 |
| Results tracked | 13 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Granite 3.1 8b Instruct: 33.9 (#232), Mercury 2: 33.2 (#250)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Coding | 1186 | 1391 |
| LMArena WebDev | — | 1170 |
| SciCode | — | 38.7% |
| WeirdML | — | 43.2% |
| ALE-Bench | — | 785.58 |
Agentic & Tool Use Not comparable
Granite 3.1 8b Instruct: 23.9 (#113), Mercury 2: —
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 27.1% | — |
Reasoning Mercury 2 leads
Granite 3.1 8b Instruct: 21.2 (#204), Mercury 2: 23.0 (#170)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Hard Prompts | 1146 | 1362 |
| CritPt | — | 0.8% |
Math Not comparable
Granite 3.1 8b Instruct: 30.4 (#210), Mercury 2: —
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Math | 1152 | — |
Knowledge Mercury 2 leads
Granite 3.1 8b Instruct: 30.6 (#218), Mercury 2: 35.6 (#170)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Expert | 1142 | 1358 |
| Vectara Hallucination Rate | — | 12.3% |
Multilingual Mercury 2 leads
Granite 3.1 8b Instruct: 31.1 (#255), Mercury 2: 46.8 (#153)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1099 | 1331 |
| LMArena Chinese | 1145 | 1416 |
| LMArena Russian | 1092 | 1303 |
Instruction Following Mercury 2 leads
Granite 3.1 8b Instruct: 58.9 (#254), Mercury 2: 70.4 (#161)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1131 | 1328 |
Long Context Mercury 2 leads
Granite 3.1 8b Instruct: 35.2 (#236), Mercury 2: 40.4 (#149)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1162 | 1329 |
Writing & Preference Mercury 2 leads
Granite 3.1 8b Instruct: 35.7 (#261), Mercury 2: 54.0 (#151)
| Benchmark | Granite 3.1 8b Instruct | Mercury 2 |
|---|---|---|
| LMArena Text | 1150 | 1355 |
| LMArena Creative Writing | 1129 | 1289 |
| LMArena Multi-Turn | 1108 | 1358 |
Frequently asked questions
Is Granite 3.1 8b Instruct better than Mercury 2?
Mercury 2 is the stronger model overall, scoring 38.8 to 31.8 on the Noometry Index.
Is Granite 3.1 8b Instruct or Mercury 2 better for coding?
They score almost the same on coding (33.9 vs 33.2); test both on your own repository before choosing.
How many benchmarks do Granite 3.1 8b Instruct and Mercury 2 share?
11 benchmarks have published results for both models. Granite 3.1 8b Instruct has 13 scored results on Noometry and Mercury 2 has 17.