See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
GPT-5.6 Sol
OpenAI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: Agent能力评测 +25.7 / Relative gap: 科学与综合推理 -0.9
Relative edge: 科学与综合推理 +0.9 / Relative gap: Agent能力评测 -25.7
Method: for each model and benchmark, the chart first averages all scores in the current mode scope instead of taking the best score, then averages those benchmark scores within each category. Only benchmarks with at least two selected models scored are included; missing values are not counted as zero.
Best overall
GPT-5.6 Sol · 77.76
Best single
GPT-5.6 Sol · GPQA Diamond 93.50
Modality coverage
Qwen3.8-Max · 3 modalities
Head to head
7
Benchmarks
6
Wins
1
Losses
+13.25
Average diff
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Complete scores for each model/mode across selected benchmarks.
7 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | GPT-5.6 Sol | Qwen3.8-Max |
|---|---|---|
GPQA Diamond 科学与综合推理 | 93.50Thinking Level · High | 92.60Thinking Level · Extra High |
DeepSWE 编程与软件工程 | 72.70Thinking Level · Extra High ｜ Tools | 56.60Thinking Level · Extra High ｜ Tools |
SWE-Bench Pro - Public 编程与软件工程 | 64.60Thinking Level · Extra High ｜ Tools | 67.70Thinking Level · Extra High ｜ Tools |
Terminal-Bench 2.1 AI Agent - 工具使用 | 88.80Thinking Level · High | 86.60Thinking Level · Extra High ｜ Tools |
Agents' Last Exam Agent能力评测 | 52.70Thinking Level · Extra High ｜ Tools | 27.00Thinking Level · Extra High ｜ Tools |
82.93Thinking Level · High | 46.34Thinking Level · Extra High | |
FrontierMath v2 数学推理 | 89.12Thinking Level · High | 74.74Thinking Level · Extra High |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | GPT-5.6 SolOpenAI | Qwen3.8-Max阿里巴巴 |
|---|---|---|
Core specsRelease | 2026-06-26 | 2026-08-03 |
Context length | 1.05M | 1M |
Parameters | — | 2.4T |
Active parameters | Not provided | 95B |
Max output | 128000 | 131072 |
MoE | No | Yes |
LicenseCode Open Source | Closed Source | Open Source · Qwen3.8-Max License |
Weights Open Source | Closed Source | Open Source · Qwen3.8-Max License |
Commercial use | 不开源 | 有条件免费商用授权 |
Local deploymentWeight size | Not provided | 约 4.89 TB（BF16，213 个 safetensors 分片） |
VRAM for weights | Not provided | ≈ 4890 GB (weights only, excludes KV cache) |
Weights | Not provided | Hugging Face |
Modality supportText Input/Output | / | / |
Image Input/Output | / | / |
Video Input/Output | Not provided | / |
ResourcesPaper / report | Previewing GPT-5.6 Sol: a next-generation model | 阿里Qwen3.8-Max模型发布：2.4万亿参数，瞄准复杂长程任务 |
Qwen3.8-Max
阿里巴巴
