#22 GPT-5.2 Chat
none- Cost
- $0.010
- Time
- 15.3s
- Tokens
- 797 tok
AI BENCHY Compare
OpenAI: GPT-5.2 Chat vs StepFun: Step 3.7 Flash
Summary
The average score is effectively tied at 8.5 vs 8.5. Step 3.7 Flash (medium) has the lower benchmark cost at $0.376 vs $0.393. GPT-5.2 Chat is faster at 7.13s vs 20.35s, with pass rates of 74.6% vs 73.0%.
Recommended modelGPT-5.2 ChatIt has the best score here (8.5), while responding about 2.9x faster than Step 3.7 Flash (medium).
Last updated at: 2026-07-14
| Metric | GPT-5.2 Chat GPT-5.2 Chat none | Step 3.7 Flash Step 3.7 Flash medium |
|---|---|---|
| Score | 8.5 | 8.5 |
| Rank | #22 | #23 |
| Reliability | 10.0 | 9.9 |
| Consistency | 8.9 | 9.3 |
| Tests Correct | ||
| Attempt pass rate | 74.6% | 73.0% |
| Flaky tests | 3 | 2 |
| Total Runs | 63 | 61 |
| Cost per result | 2.803 | 2.686 |
| Total Cost | $0.393 | $0.376 |
| Input Price | $1.750 / 1M | $0.200 / 1M |
| Output Price | $14.000 / 1M | $1.150 / 1M |
| Total Input Tokens | 34,212 | 39,981 |
| Output Tokens | 23,744 | 319,958 |
| Reasoning Tokens | 0 | 0 |
| Response Time (avg) | 7.13s | 20.35s |
| Response Time (max) | 38.52s | 113.98s |
| Response Time (total) | 149.69s | 427.42s |
Generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#23 Step 3.7 Flash
medium- Cost
- $0.006
- Time
- 46.2s
- Tokens
- 4,466 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
| Coding | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2 Chat | 8.8 | 7.8 | 88.9% | 1 | 9.82s | 7,305 | 6,731 | 0 | |
| Step 3.7 Flash | 8.8 | 7.8 | 88.9% | 1 | 27.42s | 7,437 | 44,797 | 0 |
Quick Compare
Switch Comparison Pair
Seed-2.0-LitemediumvsGPT-5.2 ChatnoneGPT-5.2 ChatnonevsGLM 5mediumGPT-5.2 ChatnonevsGrok 4.5mediumGPT-5.2 ChatnonevsGLM 5.2mediumDeepSeek V4 FlashhighvsStep 3.7 FlashmediumDeepSeek V4 FlashhighvsGPT-5.2 ChatnoneClaude Opus 4.7mediumvsGPT-5.2 ChatnoneGemini 2.5 FlashmediumvsGPT-5.2 ChatnoneGPT-5.6 TerrahighvsStep 3.7 FlashmediumGPT-5.2 ChatnonevsQwen3.7 PlusmediumClaude Opus 4.8mediumvsGPT-5.2 ChatnoneNemotron 3 Ultra 550b A55bmediumFree AvailablevsGPT-5.2 Chatnone