Model comparison
DeepSeek V4.1 Flash vs Grok 4.5
Grok 4.5 is the stronger model overall, scoring 55.0 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 11× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
DeepSeekSummary
- They share 36 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 6 categories and Grok 4.5 in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Grok 4.5 leads 44.4 to 31.2.
- The biggest single-benchmark swing is Surface Evolver Bench: 46.3% for DeepSeek V4.1 Flash and 74.4% for Grok 4.5.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $2 / $6 for Grok 4.5.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 500K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4.1 Flash | Grok 4.5 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 52.8 | 55.0 |
| Released | 2026-09-09 | 2026-07-08 |
| Weights | Open | Proprietary |
| Context window | 1M | 500K |
| Max output | 393K | 500K |
| Input $ / M tokens | $0.15 | $2 |
| Output $ / M tokens | $0.60 | $6 |
| Results tracked | 37 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek V4.1 Flash: 52.9 (#32), Grok 4.5: 52.2 (#35)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| LMArena WebDev | 1619 | 1553 |
| SciCode | 51.9% | 54.1% |
| LMArena Coding | 1506 | 1474 |
| ALE-Bench | 1,092 | 1,309 |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.4% |
| WeirdML | — | 46.4% |
Agentic & Tool Use Grok 4.5 leads
DeepSeek V4.1 Flash: 31.2 (#69), Grok 4.5: 44.4 (#17)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| APEX-Agents | 39.5% | 56.2% |
| GDP.pdf | 19.8% | 14% |
| τ²-bench Banking | — | 47.9% |
| PostTrainBench | — | 23.4% |
| GBAEval | — | 65.4% |
| LMArena Search | — | 1213 |
| Vending-Bench 2 | — | 3,887 |
Reasoning Grok 4.5 leads
DeepSeek V4.1 Flash: 50.2 (#36), Grok 4.5: 56.1 (#25)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| NYT Connections (extended) | 89.6% | 79.9% |
| CritPt | 14.3% | 15.4% |
| LMArena Hard Prompts | 1483 | 1462 |
| DTBench | 89.9% | 96.5% |
| LMCA | 47% | 45.2% |
| Surface Evolver Bench | 46.3% | 74.4% |
| Epoch Capabilities Index | 154.9 | 153.92 |
| ARC-AGI-2 | — | 52.6% |
| SimpleBench | — | 70% |
| Kagi LLM Benchmark | — | 83.5% |
| ARC-AGI-1 | — | 87.2% |
| Chess Puzzles | — | 36% |
| Mystery Game Puzzles | 43% | — |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), Grok 4.5: 60.9 (#35)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 57.2% |
| FrontierMath Tier 4 | 26.8% | 24.4% |
| OTIS Mock AIME 2024-2025 | 98.3% | 97.8% |
| ProofBench | 54% | 31% |
| LMArena Math | 1477 | 1459 |
Knowledge Grok 4.5 leads
DeepSeek V4.1 Flash: 57.9 (#38), Grok 4.5: 62.3 (#24)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| GPQA Diamond | 89.8% | 93.4% |
| LMArena Expert | 1506 | 1466 |
| SimpleQA Verified | — | 48.3% |
Multimodal DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 39.1 (#61), Grok 4.5: 37.6 (#72)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| LMArena Vision | 1277 | 1288 |
| Furniture Assembly | 34.2% | 22.5% |
| Blueprint-Bench 2 | — | 27.3% |
| LMArena Document | — | 1452 |
Multilingual Too close to call
DeepSeek V4.1 Flash: 55.0 (#35), Grok 4.5: 54.4 (#42)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1448 | 1440 |
| LMArena Chinese | 1497 | 1496 |
| LMArena French | 1452 | 1456 |
| LMArena German | 1484 | 1446 |
| LMArena Japanese | 1412 | 1428 |
| LMArena Korean | 1452 | 1404 |
| LMArena Russian | 1471 | 1448 |
| LMArena Spanish | 1459 | 1450 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), Grok 4.5: 76.0 (#48)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1446 |
Long Context Too close to call
DeepSeek V4.1 Flash: 45.2 (#47), Grok 4.5: 44.8 (#56)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1475 | 1463 |
Writing & Preference Too close to call
DeepSeek V4.1 Flash: 65.4 (#48), Grok 4.5: 65.8 (#42)
| Benchmark | DeepSeek V4.1 Flash | Grok 4.5 |
|---|---|---|
| LMArena Text | 1462 | 1448 |
| LMArena Creative Writing | 1435 | 1442 |
| EQ-Bench Creative Writing | 1540 | 1579 |
| LMArena Multi-Turn | 1457 | 1456 |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Grok 4.5?
Grok 4.5 is the stronger model overall, scoring 55.0 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 11× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4.1 Flash or Grok 4.5?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Grok 4.5 lists at $2 and $6.
Is DeepSeek V4.1 Flash or Grok 4.5 better for coding?
They score almost the same on coding (52.9 vs 52.2); test both on your own repository before choosing.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 500K.
How many benchmarks do DeepSeek V4.1 Flash and Grok 4.5 share?
36 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Grok 4.5 has 52.