GPT-6.1 Sol vs Qwen3.8
GPT-6.1 Sol is from OpenAI, released September 29, 2026. Qwen3.8 is from Alibaba, released August 3, 2026.
- LMArena Elo1484 vs 1483. GPT-6.1 Sol is 0.1% ahead on LMArena Elo.
- Input per 1M$2 vs $2. Both $2 per million input tokens.
- Output per 1M$10 vs $6. Qwen3.8 costs 40% less per million output tokens.
- Context window1.05M vs 262k. GPT-6.1 Sol takes 301% more context in one request.
Prices are the vendors’ published API rates; Elo is LMArena’s text board. Each figure is dated on its model page.
Interactive Comparison
Loading chart...
$2 / $10 per million tokens | $2 / $6 per million tokens | |
|---|---|---|
| Identity | ||
| Developer | OpenAI | Alibaba |
| Released | Sep 2026 | Aug 2026 |
| Status | GA | GA |
| Licence | Proprietary | Open weights under a licence named "qwen3.8-max" - NOT Apache 2.0 (the Qwen3.8-27B sibling is Apache 2.0, which is a common source of confusion) |
| Self-hostable | No | Yes - Qwen/Qwen3.8-2.4T-A95B and an FP8 checkpoint published 12 Aug 2026 |
| Cost | ||
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $4.000 / 1M tokens | $3.000 / 1M tokens |
| Input price | $2.000 / 1M tokens | $2.000 / 1M tokens |
| Output price | $10.000 / 1M tokens | $6.000 / 1M tokens |
| Cached input | $0.10 per 1M tokens | Not documented |
| Batch discount | 50% (Batch and Flex) | Not documented |
| Free tier | — | Not documented |
| Capacity | ||
| Context window | 1,050,000 tokens | 262k native (extensible to ~1.01M) |
| Max output | 128,000 tokens | 128k tokens |
| Long-context surcharge | Over 272K input tokens: 2x input/cache rates and 1.5x output for the full request | Not documented |
| Capability | ||
| Vision in | Yes | No - text-only. Alibaba’s model card states "Multimodal inputs are not supported". The multimodal sibling is Qwen3.8-27B (Apache 2.0), which is tagged Image-Text-to-Text and does accept images and video. |
| Audio in | No | No |
| Function calling | Yes | Yes |
| Structured output | Yes | Yes |
| Extended reasoning | Yes | Yes - thinking mode is required for all interactions, not optional |
| Web search | Yes | Not documented |
| Code execution | Yes | Not documented |
| Access | ||
| API | Yes | Yes |
| Chat app | ChatGPT Work and Codex; not Chat at launch | Yes |
| Cloud marketplaces | — | Alibaba Cloud Model Studio |
| Fine-tuning | No | Yes - open weights permit it |
| Measured quality | ||
| LMArena Elo | 1484 (checked Oct 2026) | 1483 (checked Oct 2026) |
| LMArena rank | Rank 21 | Rank 22 |
| GPQA Diamond | 95 · max effort | — |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 371 | 494 |
