| Mbinu za kupinga AI | Alama | Uthabiti | Kiwango cha kupita kwa kila jaribio | Majaribio yasiyo thabiti | Majaribio sahihi | Tokeni za matokeo | Tokeni za hoja |
|---|---|---|---|---|---|---|---|
| Google: Gemini 3 Pro Preview | 10.00 | 10.00 | 100.0% | 0 | 143 | 1,107 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 13,924 | 17,208 |
AI BENCHY Compare
Google: Gemini 3 Pro Preview vs StepFun: Step 3.5 Flash
Jina la modeli:
Benchmark zimetengenezwa kutoka seti za majaribio za AI BENCHY tarehe : 2026-02-27 15:16
Muhtasari
| Kipimo | Google: Gemini 3 Pro Preview medium Toleo: Tarehe ya kutolewa haijulikani | StepFun: Step 3.5 Flash medium Toleo: Tarehe ya kutolewa haijulikani Inapatikana bure |
|---|---|---|
| Nafasi | #3 | #11 |
| Alama | 8.64 | 7.00 |
| Uthabiti | 10.00 | 8.32 |
| Gharama kwa matokeo | 0.947 | 0.000 |
| Jumla ya gharama | $0.114 | $0.000 |
| Majaribio sahihi | ||
| Majaribio yenye makosa | 2 | 5 |
| Kiwango cha kupita kwa kila jaribio | 85.7% | 73.8% |
| Majaribio yasiyo thabiti | 0 | 3 |
| Tokeni za matokeo | 1,100 | 60,502 |
| Tokeni za hoja | 5,910 | 117,044 |
Mgawanyo wa kategoria
| Uchanganuzi na uchimbaji wa data | Alama | Uthabiti | Kiwango cha kupita kwa kila jaribio | Majaribio yasiyo thabiti | Majaribio sahihi | Tokeni za matokeo | Tokeni za hoja |
|---|---|---|---|---|---|---|---|
| Google: Gemini 3 Pro Preview | 10.00 | 10.00 | 100.0% | 0 | 318 | 687 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 535 | 11,548 |
| Mahususi kwa domeni | Alama | Uthabiti | Kiwango cha kupita kwa kila jaribio | Majaribio yasiyo thabiti | Majaribio sahihi | Tokeni za matokeo | Tokeni za hoja |
|---|---|---|---|---|---|---|---|
| Google: Gemini 3 Pro Preview | 4.00 | 10.00 | 33.3% | 0 | 15 | 1,211 | |
| StepFun: Step 3.5 Flash | 4.00 | 7.21 | 44.4% | 1 | 40,942 | 74,237 |
| Ufuataji wa maagizo | Alama | Uthabiti | Kiwango cha kupita kwa kila jaribio | Majaribio yasiyo thabiti | Majaribio sahihi | Tokeni za matokeo | Tokeni za hoja |
|---|---|---|---|---|---|---|---|
| Google: Gemini 3 Pro Preview | 9.50 | 10.00 | 100.0% | 0 | 69 | 754 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 2,121 | 3,274 |
| Puzzle Solving | Alama | Uthabiti | Kiwango cha kupita kwa kila jaribio | Majaribio yasiyo thabiti | Majaribio sahihi | Tokeni za matokeo | Tokeni za hoja |
|---|---|---|---|---|---|---|---|
| Google: Gemini 3 Pro Preview | 10.00 | 10.00 | 100.0% | 0 | 231 | 1,180 | |
| StepFun: Step 3.5 Flash | 2.00 | 4.96 | 33.3% | 2 | 2,705 | 6,975 |
| Mwito wa zana | Alama | Uthabiti | Kiwango cha kupita kwa kila jaribio | Majaribio yasiyo thabiti | Majaribio sahihi | Tokeni za matokeo | Tokeni za hoja |
|---|---|---|---|---|---|---|---|
| Google: Gemini 3 Pro Preview | 10.00 | 10.00 | 100.0% | 0 | 324 | 971 | |
| StepFun: Step 3.5 Flash | 10.00 | 10.00 | 100.0% | 0 | 275 | 3,802 |
Badilisha jozi ya ulinganisho
Linganisha Google: Gemini 3 Pro Preview dhidi ya...
- Google: Gemini 3 Flash Preview medium
- Google: Gemini 3.1 Pro Preview medium
- Qwen: Qwen3.5 Plus 2026-02-15 medium
- Qwen: Qwen3.5-27B medium
- Google: Gemini 3 Flash Preview low
- OpenAI: GPT-5.3-Codex medium
- Anthropic: Claude Sonnet 4.6 medium
- Z.ai: GLM 5 medium
- Google: Gemini 3 Flash Preview none
- StepFun: Step 3.5 Flash medium Inapatikana bure
- OpenAI: GPT-5.2 medium
- Qwen: Qwen3.5-122B-A10B medium
- Anthropic: Claude Sonnet 4.6 none
- Qwen: Qwen3.5 Plus 2026-02-15 none
- xAI: Grok 4.1 Fast medium
- MoonshotAI: Kimi K2.5 medium
- ByteDance Seed: Seed-2.0-Mini medium
- Z.ai: GLM 5 none
- Anthropic: Claude Opus 4.6 medium
- Xiaomi: MiMo-V2-Flash medium
- Qwen: Qwen3.5-35B-A3B medium
- OpenAI: GPT-5 Nano medium
- Qwen: Qwen3.5-Flash medium
- OpenAI: gpt-oss-120b medium Inapatikana bure
- MiniMax: MiniMax M2.5 medium
- Qwen: Qwen3.5-122B-A10B none
- OpenAI: GPT-4o-mini none
- Qwen: Qwen3.5-27B none
- Qwen: Qwen3.5-35B-A3B none
- Z.ai: GLM 4.7 Flash medium
- MoonshotAI: Kimi K2.5 none
- Qwen: Qwen3 Coder Next none
- Z.ai: GLM 4.7 Flash none
- Qwen: Qwen3.5-Flash none
- Qwen: Qwen3 Coder Next medium
- Xiaomi: MiMo-V2-Flash none
- xAI: Grok 4.1 Fast none
- LiquidAI: LFM2-24B-A2B none
Linganisha StepFun: Step 3.5 Flash dhidi ya...
- Google: Gemini 3 Flash Preview medium
- Google: Gemini 3.1 Pro Preview medium
- Google: Gemini 3 Pro Preview medium
- Qwen: Qwen3.5 Plus 2026-02-15 medium
- Qwen: Qwen3.5-27B medium
- Google: Gemini 3 Flash Preview low
- OpenAI: GPT-5.3-Codex medium
- Anthropic: Claude Sonnet 4.6 medium
- Z.ai: GLM 5 medium
- Google: Gemini 3 Flash Preview none
- OpenAI: GPT-5.2 medium
- Qwen: Qwen3.5-122B-A10B medium
- Anthropic: Claude Sonnet 4.6 none
- Qwen: Qwen3.5 Plus 2026-02-15 none
- xAI: Grok 4.1 Fast medium
- MoonshotAI: Kimi K2.5 medium
- ByteDance Seed: Seed-2.0-Mini medium
- Z.ai: GLM 5 none
- Anthropic: Claude Opus 4.6 medium
- Xiaomi: MiMo-V2-Flash medium
- Qwen: Qwen3.5-35B-A3B medium
- OpenAI: GPT-5 Nano medium
- Qwen: Qwen3.5-Flash medium
- OpenAI: gpt-oss-120b medium Inapatikana bure
- MiniMax: MiniMax M2.5 medium
- Qwen: Qwen3.5-122B-A10B none
- OpenAI: GPT-4o-mini none
- Qwen: Qwen3.5-27B none
- Qwen: Qwen3.5-35B-A3B none
- Z.ai: GLM 4.7 Flash medium
- MoonshotAI: Kimi K2.5 none
- Qwen: Qwen3 Coder Next none
- Z.ai: GLM 4.7 Flash none
- Qwen: Qwen3.5-Flash none
- Qwen: Qwen3 Coder Next medium
- Xiaomi: MiMo-V2-Flash none
- xAI: Grok 4.1 Fast none
- LiquidAI: LFM2-24B-A2B none