Model comparison
DeepSeek-V3.2-Exp vs Grok 4.3
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 43.6 on the Noometry Index.
Last verified . 30 shared benchmarks.
DeepSeekSummary
- They share 30 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 6 categories and Grok 4.3 in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.3 leads 34.9 to 22.0.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 55.2% for Grok 4.3.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 164K.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.2-Exp | Grok 4.3 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 44.9 | 43.6 |
| Released | 2025-09-29 | 2026-04-17 |
| Weights | Open | Proprietary |
| Context window | 164K | 1M |
| Max output | 66K | 30K |
| Input $ / M tokens | $0.26 | $1.25 |
| Output $ / M tokens | $0.38 | $2.50 |
| Results tracked | 48 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 46.4 (#61), Grok 4.3: 41.3 (#117)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| LMArena WebDev | 1362 | 1357 |
| SciCode | 38.9% | 47.3% |
| WeirdML | 39.5% | 49.9% |
| LMArena Coding | 1454 | 1415 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| ALE-Bench | — | 944.17 |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 37.9 (#31), Grok 4.3: 27.6 (#93)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| Vending-Bench 2 | 1,034 | 35.26 |
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |
Reasoning Grok 4.3 leads
DeepSeek-V3.2-Exp: 22.0 (#185), Grok 4.3: 34.9 (#60)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| NYT Connections (extended) | 36.7% | 55.2% |
| CritPt | 2.9% | 8% |
| Chess Puzzles | 14% | 25% |
| LMArena Hard Prompts | 1433 | 1397 |
| DTBench | 87.7% | 90.7% |
| LMCA | 29.1% | 38.3% |
| Epoch Capabilities Index | 146.28 | 149.13 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
| ForecastBench | — | 60.3 |
Math Grok 4.3 leads
DeepSeek-V3.2-Exp: 40.8 (#75), Grok 4.3: 45.8 (#60)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 93.3% |
| ProofBench | 8% | 11% |
| LMArena Math | 1435 | 1388 |
| FrontierMath (Tiers 1-3) | — | 42.8% |
| FrontierMath Tier 4 | — | 14.6% |
| MathArena Final-Answer Competitions | 57.7% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
DeepSeek-V3.2-Exp: 51.6 (#61), Grok 4.3: 52.5 (#57)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 83.4% | 88.8% |
| LMArena Expert | 1436 | 1385 |
| SimpleQA Verified | — | 33.2% |
| Vectara Hallucination Rate | 5.3% | — |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, Grok 4.3: 31.6 (#96)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| LMArena Vision | — | 1231 |
| Blueprint-Bench 2 | — | 0% |
Multilingual DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 52.4 (#85), Grok 4.3: 50.7 (#114)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1409 | 1385 |
| LMArena Chinese | 1461 | 1423 |
| LMArena French | 1435 | 1412 |
| LMArena German | 1440 | 1395 |
| LMArena Japanese | 1377 | 1379 |
| LMArena Korean | 1372 | 1357 |
| LMArena Russian | 1423 | 1399 |
| LMArena Spanish | 1441 | 1402 |
Instruction Following DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 74.7 (#89), Grok 4.3: 72.3 (#135)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1413 | 1366 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), Grok 4.3: 42.5 (#118)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1428 | 1393 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 62.5 (#73), Grok 4.3: 58.6 (#112)
| Benchmark | DeepSeek-V3.2-Exp | Grok 4.3 |
|---|---|---|
| LMArena Text | 1425 | 1397 |
| LMArena Creative Writing | 1403 | 1380 |
| LMArena Multi-Turn | 1427 | 1406 |
| EQ-Bench Creative Writing | 1515 | — |
| EQ-Bench 4 | — | 1075 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Grok 4.3?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 43.6 on the Noometry Index.
Which is cheaper, DeepSeek-V3.2-Exp or Grok 4.3?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.
Is DeepSeek-V3.2-Exp or Grok 4.3 better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 41.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and Grok 4.3 share?
30 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and Grok 4.3 has 40.