AI BENCHY Compare
DeepSeek: DeepSeek V3.2 vs Qwen: Qwen3 Coder Next
Last updated at: 2026-03-03
| Metric | DeepSeek: DeepSeek V3.2 none Release: 2025-12-01 | Qwen: Qwen3 Coder Next medium Release: 2026-02-03 |
|---|---|---|
| Rank | #30 | #39 |
| Avg Score | 4.70 | 3.50 |
| Consistency | 8.19 | 8.43 |
| Cost per result | 0.132 | 0.135 |
| Total Cost | $0.007 | $0.005 |
| Tests Correct | 5/14 | 3/14 |
| Attempt pass rate | 47.6% | 31.0% |
| Flaky tests | 3 | 3 |
| Output Tokens | 4,869 | 2,569 |
| Reasoning Tokens | 0 | 0 |
Category Breakdown
| Anti-AI Tricks | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 1.00 | 9.68 | 0.0% | 0 | 1,411 | 0 | |
| Qwen: Qwen3 Coder Next | 1.33 | 7.49 | 22.2% | 1 | 1,246 | 0 |
| Data parsing and extraction | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 5.38 | 5.81 | 66.7% | 1 | 1,710 | 0 | |
| Qwen: Qwen3 Coder Next | 5.50 | 10.00 | 50.0% | 0 | 276 | 0 |
| Domain specific | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 1.00 | 7.21 | 22.2% | 1 | 24 | 0 | |
| Qwen: Qwen3 Coder Next | 4.00 | 10.00 | 33.3% | 0 | 24 | 0 |
| Instructions following | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 8.00 | 9.99 | 50.0% | 0 | 66 | 0 | |
| Qwen: Qwen3 Coder Next | 4.50 | 6.88 | 16.7% | 1 | 67 | 0 |
| Puzzle Solving | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 7.67 | 7.49 | 88.9% | 1 | 1,136 | 0 | |
| Qwen: Qwen3 Coder Next | 1.00 | 7.28 | 11.1% | 1 | 701 | 0 |
| Tool Calling | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|
| DeepSeek: DeepSeek V3.2 | 10.00 | 10.00 | 100.0% | 0 | 522 | 0 | |
| Qwen: Qwen3 Coder Next | 10.00 | 10.00 | 100.0% | 0 | 255 | 0 |
Quick Compare
Switch Comparison Pair
Qwen3 Coder NextmediumvsGLM 4.7 FlashnoneQwen3 Coder NextmediumvsMiMo-V2-FlashnoneQwen3 Coder NextmediumvsGrok 4.1 FastnoneDeepSeek V3.2nonevsGLM 4.7 FlashmediumKimi K2.5nonevsQwen3 Coder NextmediumTrinity Large Preview (free)noneFree AvailablevsQwen3 Coder NextmediumDeepSeek V3.2nonevsMiniMax M2.5mediumDeepSeek V3.2nonevsgpt-oss-120bmediumFree AvailableLFM2-24B-A2BnonevsQwen3 Coder NextmediumDeepSeek V3.2nonevsQwen3.5-FlashmediumDeepSeek V3.2nonevsGPT-5 NanomediumDeepSeek V3.2nonevsQwen3.5-35B-A3Bmedium