- Rank
- #54
- Total Output Tokens
- 80,538
- Response Time (avg)
- 31.29s
- Total Cost
- $1.540
GPT-5.2 (medium) vs Qwen3.8 Max (0902) (low)
GPT-5.2 (medium) leads on average score with 8.7 vs 8.6. Qwen3.8 Max (0902) (low) has the lower benchmark cost at $1.131 vs $1.540. Qwen3.8 Max (0902) (low) is faster at 29.07s vs 31.29s, with pass rates of 73.9% vs 85.5%.
Compared models
- Rank
- #63
- Total Output Tokens
- 88,923
- Response Time (avg)
- 29.07s
- Total Cost
- $1.131
Recommended model
Qwen3.8 Max (0902) (low)
It offers the best overall trade-off: a competitive score (8.6), lower cost than GPT-5.2 (medium), and balanced response time.
Detailed comparison
| Metric | GPT-5.2 GPT-5.2 medium | Qwen3.8 Max (0902) Qwen3.8 Max (0902) low |
|---|---|---|
| Score | 8.7 | 8.6 |
| Rank | #54 | #63 |
| Reliability | 10.0 | 10.0 |
| Consistency | 8.5 | 9.0 |
| Attempts | 69/69 | 69/69 |
| Tests Correct | ||
| Attempt pass rate | 73.9% | 85.5% |
| Flaky tests | 4 | 3 |
| Total Runs | 69 | 69 |
| Cost per result | 10.261 | 6.283 |
| Total Cost | $1.540 | $1.131 |
| Input Price | $1.750 / 1M | $2.000 / 1M |
| Output Price | $14.000 / 1M | $6.000 / 1M |
| Total Input Tokens | 235,207 | 298,682 |
| Output Tokens | 11,065 | 8,374 |
| Reasoning Tokens | 69,473 | 80,549 |
| Response Time (avg) | 31.29s | 29.07s |
| Response Time (max) | 116.86s | 207.79s |
| Response Time (total) | 531.98s | 668.66s |
| Parameters | ~1.2T total (~80B active) | 2.4T total (~100B active) |
| Availability | Closed | Closed |
Model generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#54 GPT-5.2
medium- Cost
- $0.047
- Time
- 49.2s
- Tokens
- 3,396 tok
#63 Qwen3.8 Max (0902)
low- Cost
- $0.014
- Time
- 41.2s
- Tokens
- 2,421 tok
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Category Breakdown
| Agentic | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 10.0 | 10.0 | 100.0% | 0 | 75.85s | 130,194 | 1,151 | 8,135 | |
| Qwen3.8 Max (0902) | 6.1 | 3.1 | 66.7% | 1 | 136.04s | 194,728 | 2,052 | 11,006 |
| Anti-AI Tricks | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 6.5 | 8.0 | 58.3% | 1 | 7.81s | 606 | 567 | 2,002 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 4.72s | 984 | 212 | 1,854 |
| Coding | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 10.0 | 10.0 | 100.0% | 0 | 22.73s | 7,302 | 511 | 11,912 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 37.42s | 8,127 | 485 | 17,404 |
| Combined | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 10.0 | 10.0 | 100.0% | 0 | 58.50s | 82,056 | 7,304 | 14,693 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 113.88s | 74,767 | 4,562 | 29,335 |
| Data parsing and extraction | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 10.0 | 10.0 | 100.0% | 0 | 3.15s | 7,140 | 234 | 420 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 9.94s | 7,938 | 261 | 2,254 |
| Domain specific | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 5.9 | 7.2 | 55.6% | 1 | 97.33s | 482 | 42 | 26,812 | |
| Qwen3.8 Max (0902) | 5.3 | 7.2 | 44.4% | 1 | 34.93s | 1,014 | 45 | 13,881 |
| General Intelligence | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 3.7 | 9.7 | 0.0% | 0 | 4.32s | 477 | 162 | 269 | |
| Qwen3.8 Max (0902) | 6.1 | 3.1 | 66.7% | 1 | 8.91s | 594 | 120 | 820 |
| Instructions following | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 9.9 | 10.0 | 100.0% | 0 | 3.12s | 660 | 94 | 614 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 4.32s | 855 | 90 | 866 |
| Puzzle Solving | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 7.5 | 7.3 | 77.8% | 1 | 5.80s | 642 | 735 | 924 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 4.49s | 930 | 275 | 1,248 |
| Tool Calling | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 4.7 | 1.6 | 66.7% | 1 | 10.30s | 5,453 | 239 | 469 | |
| Qwen3.8 Max (0902) | 10.0 | 10.0 | 100.0% | 0 | 4.73s | 8,463 | 249 | 126 |
| Trivia | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 | 3.0 | 10.0 | 0.0% | 0 | 28.18s | 195 | 26 | 3,223 | |
| Qwen3.8 Max (0902) | 3.0 | 10.0 | 0.0% | 0 | 13.28s | 282 | 23 | 1,755 |
Quick Compare
Switch Comparison Pair
GPT-5.2mediumvsQwen3.8 27BhighFree AvailableQwen3.8 Max (0902)lowvsGrok 4.6highClaude Sonnet 5.5highvsQwen3.8 Max (0902)lowClaude Sonnet 5.5xhighvsQwen3.8 Max (0902)lowDeepSeek V4.1 FlashmaxvsGPT-5.2mediumMuse Spark 1.2lowvsGPT-5.2mediumClaude Sonnet 5mediumvsQwen3.8 Max (0902)lowDeepSeek V4.1 FlashmaxvsQwen3.8 Max (0902)lowDeepSeek V4.1 FlashhighvsGPT-5.2mediumMuse Spark 1.1lowvsGPT-5.2mediumClaude Sonnet 5.5xhighvsGPT-5.2mediumClaude Sonnet 5.5highvsGPT-5.2medium