Model Comparison
Qwen3 Max Thinking (Preview)
vs. Seed-OSS-36B-Instruct
Comparing 2 AI models · 10 benchmarks · Alibaba, ByteDance Seed
Recommended Pick
Al Qwen3 Max Thinking (Preview) 12 metric wins Strongest on: Throughput, Latency, Reasoning
Best Value
Seed-OSS-36B-Instruct
100.0 value score
55.3 reasoning / $0.30/1M
Lowest Price
Seed-OSS-36B-Instruct
$0.21/1M input price
Best Reasoning
Qwen3 Max Thinking (Preview)
57.3 reasoning score
Blends available reasoning benchmarks
Best for Coding
Qwen3 Max Thinking (Preview)
24.5 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Best value
Seed-OSS-36B-Instruct has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Seed-OSS-36B-Instruct is 5.7x cheaper on input tokens than Qwen3 Max Thinking (Preview).
Speed gap
Qwen3 Max Thinking (Preview) generates about 1.9x as many tokens per second as Seed-OSS-36B-Instruct.
Reasoning gap
Qwen3 Max Thinking (Preview) leads Seed-OSS-36B-Instruct by 2.1 points on reasoning.
Coding gap
Qwen3 Max Thinking (Preview) leads Seed-OSS-36B-Instruct by 7.8 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Qwen3 Max Thinking (Preview)
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Seed-OSS-36B-Instruct
ByteDance Seed
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Top Pick Al Qwen3 Max Thinking (Preview) | By Seed-OSS-36B-Instruct |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $1.20/1M | $0.21/1M |
| Output Cost | $6.00/1M | $0.57/1M |
| Blended (3:1) | $2.40/1M | $0.30/1M |
| Specifications | ||
| Organization | Alibaba | ByteDance Seed |
| Release Date | Nov 3, 2025 | Aug 20, 2025 |
| Performance & Speed | ||
| Throughput | 54.9 tok/s | 29.3 tok/s |
| TTFT | 1727ms | 1574ms |
| Latency | 38129ms | 69874ms |
| Composite Indices | ||
| Value Score | 13.0 | 100.0 |
| Reasoning Score | 57.3 | 55.3 |
| Intelligence | 32.5 | 25.2 |
| Coding | 24.5 | 16.7 |
| Math | 82.3 | 84.7 |
| Standard Benchmarks | ||
| GPQA | 77.6% | 72.6% |
| MMLU Pro | 82.4% | 81.5% |
| HLE | 12.0% | 9.1% |
| LiveCodeBench | 53.5% | 76.5% |
| AIME 2025 | 82.3% | 84.7% |
| SciCode | 38.7% | 36.5% |
| LCR | 57.7% | 57.7% |
| IFBench | 53.8% | 41.9% |
| TAU-bench v2 | 83.6% | 49.4% |
| TerminalBench Hard | 17.4% | 6.8% |
Key Takeaways
Seed-OSS-36B-Instruct offers the best value at $0.21/1M, making it ideal for high-volume applications and cost-conscious projects.
Qwen3 Max Thinking (Preview) has the strongest reasoning profile with a 57.3 reasoning score, combining the available reasoning-heavy benchmarks.
Qwen3 Max Thinking (Preview) reaches a 24.5 coding index, making it the top choice for software development and code generation tasks.
All models support context windows of ∞+ tokens, suitable for processing lengthy documents and maintaining extended conversations.
When to Choose Each Model
Qwen3 Max Thinking (Preview)
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
Seed-OSS-36B-Instruct
- Cost-sensitive applications
- High-volume processing
