Model comparison
DeepSeek-V3.2-Exp vs Hy3
DeepSeek-V3.2-Exp and Hy3 score almost the same on the Noometry Index (44.3 vs 44.2), so choose on price, context window or the category you care about most.
Last verified . 19 shared benchmarks.
DeepSeek
TencentSummary
- They share 19 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 4 categories and Hy3 in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.7 to 40.8.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
- Hy3 accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | Hy3 | |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 44.3 | 44.2 |
| Released | 2025-09-29 | 2026-07-06 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.26 | $0.0825 |
| Output $ / M tokens | $0.38 | $0.33 |
| Results tracked | 49 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.2-Exp: 46.5 (#65), Hy3: 46.8 (#63)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena WebDev | 1362 | 1508 |
| LMArena Coding | 1454 | 1464 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| SciCode | 38.9% | — |
| WeirdML | 39.5% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.2-Exp: 32.7 (#59), Hy3: —
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| APEX-Agents | 21.3% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| Vending-Bench 2 | 1,034 | — |
Reasoning Hy3 leads
DeepSeek-V3.2-Exp: 22.1 (#208), Hy3: 26.1 (#136)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| NYT Connections (extended) | 36.7% | 41.2% |
| LMArena Hard Prompts | 1434 | 1447 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| CritPt | 2.9% | — |
| Chess Puzzles | 14% | — |
| Thematic Generalization | 65% | — |
| DTBench | 87.7% | — |
| LMCA | 29.1% | — |
| Epoch Capabilities Index | 146.27 | — |
Math DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 41.7 (#87), Hy3: 40.1 (#93)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena Math | 1435 | 1475 |
| MathArena Final-Answer Competitions | 57.7% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| ProofBench | 8% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.7 (#66), Hy3: 40.8 (#114)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena Expert | 1436 | 1460 |
| GPQA Diamond | 83.4% | — |
| Vectara Hallucination Rate | 5.3% | — |
Multilingual Hy3 leads
DeepSeek-V3.2-Exp: 52.2 (#90), Hy3: 53.5 (#65)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena Non-English | 1409 | 1426 |
| LMArena Chinese | 1461 | 1493 |
| LMArena French | 1433 | 1461 |
| LMArena German | 1440 | 1439 |
| LMArena Japanese | 1374 | 1392 |
| LMArena Korean | 1371 | 1395 |
| LMArena Russian | 1424 | 1432 |
| LMArena Spanish | 1440 | 1456 |
Instruction Following Too close to call
DeepSeek-V3.2-Exp: 74.5 (#93), Hy3: 75.1 (#70)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1413 | 1426 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#16), Hy3: 44.1 (#75)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena Longer Query | 1428 | 1442 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference Too close to call
DeepSeek-V3.2-Exp: 62.4 (#77), Hy3: 62.2 (#81)
| Benchmark | DeepSeek-V3.2-Exp | Hy3 |
|---|---|---|
| LMArena Text | 1425 | 1439 |
| LMArena Creative Writing | 1403 | 1402 |
| LMArena Multi-Turn | 1427 | 1436 |
| EQ-Bench Creative Writing | 1515 | — |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Hy3?
DeepSeek-V3.2-Exp and Hy3 score almost the same on the Noometry Index (44.3 vs 44.2), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.2-Exp or Hy3?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.
Is DeepSeek-V3.2-Exp or Hy3 better for coding?
They score almost the same on coding (46.5 vs 46.8); test both on your own repository before choosing.
Which has the bigger context window?
Hy3 does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and Hy3 share?
19 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Hy3 has 19.