AA-Briefcase
1.1 / combined Elo ↗- DeepSeek V4.1 Flash:
- GPT-6 Astra:
Compare models
Find the right model, side by side.
Updated: Data as of:
First model
Second model
Model specifications
| Specifications |  DeepSeek V4.1 FlashDeepSeek |  GPT-6 AstraOpenAI |
|---|---|---|
| Provider | ||
| Context sizetokens | 1,000,000 | 1,000,000 |
| Cutoff date | Not reported | 2026-04-30 |
| Input costUSD / 1M tokens | $0.30 | $10.00 |
| Output costUSD / 1M tokens | $1.20 | $50.00 |
| Max outputtokens | Not reported | 128,000 |
| LatencyTTFT · seconds | 0.63 | 365.127 |
| Speedtokens / sec | 244.981 | 70.564 |
Benchmark performance
7 shared benchmark comparisons. Versions and setups matched.
DeepSeek V4.1 FlashGPT-6 Astra
AutomationBench-AA
1.0.6 / AA private 657 ↗Tools enabled- DeepSeek V4.1 Flash:
- GPT-6 Astra:
- DeepSeek V4.1 Flash:
- GPT-6 Astra:
GDPval-AA
2.1 ↗- DeepSeek V4.1 Flash:
- GPT-6 Astra:
HLE-Diamond
1,000 questions / no tools ↗No tools- DeepSeek V4.1 Flash:
- GPT-6 Astra:
Humanity’s Last Exam
Stable HLE — text-only subset ↗No tools- DeepSeek V4.1 Flash:
- GPT-6 Astra:
Terminal-Bench
4.0 / mini-swe-agent ↗mini-swe-agent- DeepSeek V4.1 Flash:
- GPT-6 Astra:
