Model comparison
DeepSeek-V3.2-Speciale vs Grok 4 Fast
DeepSeek-V3.2-Speciale and Grok 4 Fast score almost the same on the Noometry Index (39.1 vs 39.0), so choose on price, context window or the category you care about most.
Last verified . 1 shared benchmarks.
DeepSeekSummary
- They share 1 benchmark with published results for both. DeepSeek-V3.2-Speciale scores higher in 2 categories and Grok 4 Fast in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok 4 Fast leads 60.2 to 46.1.
- DeepSeek-V3.2-Speciale has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.2-Speciale | Grok 4 Fast | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 39.1 | 39.0 |
| Released | 2025-12-01 | 2025-09-19 |
| Weights | Open | Proprietary |
| Context window | 128K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.58 | — |
| Output $ / M tokens | $1.68 | — |
| Results tracked | 3 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 39.9 (#136), Grok 4 Fast: 32.3 (#264)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| WeirdML | 46.7% | 42.9% |
| LMArena WebDev | — | 1160 |
| LMArena Coding | — | 1429 |
Agentic & Tool Use Not comparable
DeepSeek-V3.2-Speciale: —, Grok 4 Fast: 29.3 (#80)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| τ²-bench Banking | — | 15.7% |
| Cybench | — | 30% |
| LMArena Search | — | 1171 |
Reasoning DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 32.1 (#65), Grok 4 Fast: 22.5 (#179)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| ARC-AGI-2 | — | 5.3% |
| SimpleBench | 52.6% | — |
| Kagi LLM Benchmark | — | 66.1% |
| ARC-AGI-1 | — | 48.5% |
| LMArena Hard Prompts | — | 1413 |
| DTBench | — | 82.7% |
| Epoch Capabilities Index | — | 144.21 |
| ForecastBench | — | 60.5 |
Math Not comparable
DeepSeek-V3.2-Speciale: —, Grok 4 Fast: 36.1 (#126)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| LMArena Math | — | 1419 |
Knowledge Not comparable
DeepSeek-V3.2-Speciale: —, Grok 4 Fast: 31.5 (#211)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| Vectara Hallucination Rate | — | 19.7% |
| LMArena Expert | — | 1411 |
Multilingual Not comparable
DeepSeek-V3.2-Speciale: —, Grok 4 Fast: 51.5 (#106)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| LMArena Non-English | — | 1397 |
| LMArena Chinese | — | 1458 |
| LMArena French | — | 1433 |
| LMArena German | — | 1383 |
| LMArena Japanese | — | 1358 |
| LMArena Korean | — | 1359 |
| LMArena Russian | — | 1387 |
| LMArena Spanish | — | 1414 |
Instruction Following Not comparable
DeepSeek-V3.2-Speciale: —, Grok 4 Fast: 73.4 (#116)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| LMArena Instruction Following | — | 1388 |
Long Context Not comparable
DeepSeek-V3.2-Speciale: —, Grok 4 Fast: 63.2 (#3)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| Fiction.LiveBench | — | 94.4% |
| LMArena Longer Query | — | 1415 |
Writing & Preference Grok 4 Fast leads
DeepSeek-V3.2-Speciale: 46.1 (#217), Grok 4 Fast: 60.2 (#96)
| Benchmark | DeepSeek-V3.2-Speciale | Grok 4 Fast |
|---|---|---|
| LMArena Text | — | 1407 |
| LMArena Creative Writing | — | 1388 |
| EQ-Bench Creative Writing | 1276 | — |
| LMArena Multi-Turn | — | 1414 |
Frequently asked questions
Is DeepSeek-V3.2-Speciale better than Grok 4 Fast?
DeepSeek-V3.2-Speciale and Grok 4 Fast score almost the same on the Noometry Index (39.1 vs 39.0), so choose on price, context window or the category you care about most.
Is DeepSeek-V3.2-Speciale or Grok 4 Fast better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 39.9 versus 32.3 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Speciale and Grok 4 Fast share?
1 benchmark has published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and Grok 4 Fast has 30.