Model comparison
DeepSeek V4 Pro vs Solar Pro4
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.1 on the Noometry Index. Solar Pro4 costs 1.9× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
DeepSeek
UpstageSummary
- They share 18 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Solar Pro4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 28.5.
- Solar Pro4 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
- DeepSeek V4 Pro accepts more context: 1M tokens versus 524K.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Pro | Solar Pro4 | |
|---|---|---|
| Provider | DeepSeek | Upstage |
| Noometry Index | 54.3 | 42.1 |
| Released | 2026-04-24 | 2026-08-06 |
| Weights | Open | Proprietary |
| Context window | 1M | 524K |
| Max output | 393K | 131K |
| Input $ / M tokens | $0.66 | $0.30 |
| Output $ / M tokens | $1.98 | $1.20 |
| Results tracked | 48 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Solar Pro4: 40.1 (#149)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena WebDev | 1582 | 1371 |
| LMArena Coding | 1470 | 1437 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Solar Pro4: —
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Solar Pro4: 28.5 (#104)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1399 |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| DTBench | 93.9% | — |
| LMCA | 45.5% | — |
| Surface Evolver Bench | 40% | — |
| Epoch Capabilities Index | 155.31 | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Solar Pro4: 38.8 (#128)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Math | 1455 | 1416 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| OTIS Mock AIME 2024-2025 | 98.6% | — |
| ProofBench | 50% | — |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Solar Pro4: 39.8 (#129)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Expert | 1464 | 1427 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Solar Pro4: 48.7 (#139)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Non-English | 1439 | 1361 |
| LMArena Chinese | 1486 | 1415 |
| LMArena French | 1472 | 1397 |
| LMArena German | 1458 | 1364 |
| LMArena Japanese | 1445 | 1309 |
| LMArena Korean | 1447 | 1382 |
| LMArena Russian | 1453 | 1360 |
| LMArena Spanish | 1458 | 1401 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Solar Pro4: 72.7 (#132)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Instruction Following | 1448 | 1377 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Solar Pro4: 42.1 (#130)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Longer Query | 1458 | 1381 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Solar Pro4: 56.5 (#138)
| Benchmark | DeepSeek V4 Pro | Solar Pro4 |
|---|---|---|
| LMArena Text | 1451 | 1386 |
| LMArena Creative Writing | 1446 | 1316 |
| LMArena Multi-Turn | 1467 | 1385 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Solar Pro4?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.1 on the Noometry Index. Solar Pro4 costs 1.9× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4 Pro or Solar Pro4?
Solar Pro4 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.
Is DeepSeek V4 Pro or Solar Pro4 better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Pro does, with 1M tokens against 524K.
How many benchmarks do DeepSeek V4 Pro and Solar Pro4 share?
18 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Solar Pro4 has 18.