Model comparison
DeepSeek-V3.2-Exp vs GPT-5.5 Instant
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 42.5 on the Noometry Index.
Last verified . 23 shared benchmarks.
DeepSeekSummary
- They share 23 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 6 categories and GPT-5.5 Instant in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.2-Exp leads 40.8 to 26.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 87.8% for DeepSeek-V3.2-Exp and 68.1% for GPT-5.5 Instant.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.2-Exp | GPT-5.5 Instant | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 44.9 | 42.5 |
| Released | 2025-09-29 | 2026-05-05 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.26 | — |
| Output $ / M tokens | $0.38 | — |
| Results tracked | 48 | 27 |
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), GPT-5.5 Instant: 43.7 (#73)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| SciCode | 38.9% | 48.6% |
| LMArena Coding | 1454 | 1433 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| LMArena WebDev | 1362 | — |
| SWE-bench Multilingual | 59% | — |
| WeirdML | 39.5% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.2-Exp: 37.9 (#31), GPT-5.5 Instant: —
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| Vending-Bench 2 | 1,034 | — |
Reasoning GPT-5.5 Instant leads
DeepSeek-V3.2-Exp: 22.0 (#185), GPT-5.5 Instant: 24.2 (#149)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| CritPt | 2.9% | 0% |
| Chess Puzzles | 14% | 12% |
| LMArena Hard Prompts | 1433 | 1426 |
| Epoch Capabilities Index | 146.28 | 142.48 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 36.7% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
| DTBench | 87.7% | — |
| LMCA | 29.1% | — |
Math DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 40.8 (#75), GPT-5.5 Instant: 26.1 (#241)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 68.1% |
| LMArena Math | 1435 | 1420 |
| FrontierMath (Tiers 1-3) | — | 26.3% |
| FrontierMath Tier 4 | — | 2.4% |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.6 (#61), GPT-5.5 Instant: 48.9 (#68)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| GPQA Diamond | 83.4% | 82.5% |
| LMArena Expert | 1436 | 1409 |
| Vectara Hallucination Rate | 5.3% | — |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, GPT-5.5 Instant: 39.3 (#48)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| LMArena Vision | — | 1250 |
| LMArena Document | — | 1403 |
Multilingual Too close to call
DeepSeek-V3.2-Exp: 52.4 (#85), GPT-5.5 Instant: 53.0 (#76)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| LMArena Non-English | 1409 | 1417 |
| LMArena Chinese | 1461 | 1457 |
| LMArena French | 1435 | 1428 |
| LMArena German | 1440 | 1410 |
| LMArena Japanese | 1377 | 1408 |
| LMArena Korean | 1372 | 1393 |
| LMArena Russian | 1423 | 1430 |
| LMArena Spanish | 1441 | 1430 |
Instruction Following Too close to call
DeepSeek-V3.2-Exp: 74.7 (#89), GPT-5.5 Instant: 74.4 (#94)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| LMArena Instruction Following | 1413 | 1407 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), GPT-5.5 Instant: 43.4 (#92)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| LMArena Longer Query | 1428 | 1422 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference Too close to call
DeepSeek-V3.2-Exp: 62.5 (#73), GPT-5.5 Instant: 62.0 (#81)
| Benchmark | DeepSeek-V3.2-Exp | GPT-5.5 Instant |
|---|---|---|
| LMArena Text | 1425 | 1419 |
| LMArena Creative Writing | 1403 | 1420 |
| LMArena Multi-Turn | 1427 | 1434 |
| EQ-Bench Creative Writing | 1515 | — |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than GPT-5.5 Instant?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 42.5 on the Noometry Index.
Is DeepSeek-V3.2-Exp or GPT-5.5 Instant better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 43.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Exp and GPT-5.5 Instant share?
23 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and GPT-5.5 Instant has 27.