Model comparison
Solar Pro4 vs Tulu 3 (Tülu 3) 70B
Solar Pro4 is the stronger model overall, scoring 42.1 to 33.0 on the Noometry Index.
Last verified . 11 shared benchmarks.
Upstage
Allen Institute for AI (Ai2)Summary
- They share 11 benchmarks with published results for both. Solar Pro4 scores higher in 8 categories and Tulu 3 (Tülu 3) 70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Solar Pro4 leads 38.8 to 14.2.
- Tulu 3 (Tülu 3) 70B has downloadable open weights; the other is API-only.
Side by side
| Solar Pro4 | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Upstage | Allen Institute for AI (Ai2) |
| Noometry Index | 42.1 | 33.0 |
| Released | 2026-08-06 | 2024-11-21 |
| Weights | Proprietary | Open |
| Context window | 524K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 18 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Solar Pro4 leads
Solar Pro4: 40.1 (#149), Tulu 3 (Tülu 3) 70B: 36.0 (#214)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1437 | 1235 |
| LMArena WebDev | 1371 | — |
Reasoning Solar Pro4 leads
Solar Pro4: 28.5 (#104), Tulu 3 (Tülu 3) 70B: 23.9 (#169)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1399 | 1220 |
Math Solar Pro4 leads
Solar Pro4: 38.8 (#128), Tulu 3 (Tülu 3) 70B: 14.2 (#303)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Math | 1416 | 1242 |
| OTIS Mock AIME 2024-2025 | — | 4.4% |
| MATH Level 5 | — | 42.7% |
Knowledge Solar Pro4 leads
Solar Pro4: 39.8 (#129), Tulu 3 (Tülu 3) 70B: 25.0 (#264)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | — | 46.3% |
| LMArena Expert | 1427 | — |
Multilingual Solar Pro4 leads
Solar Pro4: 48.7 (#139), Tulu 3 (Tülu 3) 70B: 39.9 (#222)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1361 | 1236 |
| LMArena Chinese | 1415 | 1249 |
| LMArena Russian | 1360 | 1246 |
| LMArena French | 1397 | — |
| LMArena German | 1364 | — |
| LMArena Japanese | 1309 | — |
| LMArena Korean | 1382 | — |
| LMArena Spanish | 1401 | — |
Instruction Following Solar Pro4 leads
Solar Pro4: 72.7 (#132), Tulu 3 (Tülu 3) 70B: 64.8 (#227)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1377 | 1233 |
Long Context Solar Pro4 leads
Solar Pro4: 42.1 (#130), Tulu 3 (Tülu 3) 70B: 37.1 (#222)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1381 | 1224 |
Writing & Preference Solar Pro4 leads
Solar Pro4: 56.5 (#138), Tulu 3 (Tülu 3) 70B: 45.6 (#223)
| Benchmark | Solar Pro4 | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1386 | 1256 |
| LMArena Creative Writing | 1316 | 1231 |
| LMArena Multi-Turn | 1385 | 1252 |
Frequently asked questions
Is Solar Pro4 better than Tulu 3 (Tülu 3) 70B?
Solar Pro4 is the stronger model overall, scoring 42.1 to 33.0 on the Noometry Index.
Is Solar Pro4 or Tulu 3 (Tülu 3) 70B better for coding?
Solar Pro4 scores higher on coding benchmarks: 40.1 versus 36.0 in the Noometry coding category.
How many benchmarks do Solar Pro4 and Tulu 3 (Tülu 3) 70B share?
11 benchmarks have published results for both models. Solar Pro4 has 18 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.