Model Comparison
Qwen3 Coder 480B A35B Instruct
vs. 105B (high)
Comparing 2 AI models · 12 benchmarks · Alibaba, Sarvam
Recommended Pick
Sa 105B (high) 9 metric wins Strongest on: Value, Input price, Output price
Best Value
105B (high)
100.0 value score
34.0 reasoning / $0.07/1M
Lowest Price
105B (high)
$0.04/1M input price
Best Reasoning
Qwen3 Coder 480B A35B Instruct
44.5 reasoning score
Blends available reasoning benchmarks
Best for Coding
Qwen3 Coder 480B A35B Instruct
24.6 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
105B (high) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
105B (high) is 7.1x cheaper on input tokens than Qwen3 Coder 480B A35B Instruct.
Speed gap
105B (high) generates about 1.3x as many tokens per second as Qwen3 Coder 480B A35B Instruct.
Reasoning gap
Qwen3 Coder 480B A35B Instruct leads 105B (high) by 10.5 points on reasoning.
Coding gap
Qwen3 Coder 480B A35B Instruct leads 105B (high) by 14.8 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Qwen3 Coder 480B A35B Instruct
Alibaba
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
105B (high)
Sarvam
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Al Qwen3 Coder 480B A35B Instruct | Top Pick Sa 105B (high) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.30/1M | $0.04/1M |
| Output Cost | $1.80/1M | $0.17/1M |
| Blended (3:1) | $0.68/1M | $0.07/1M |
| Specifications | ||
| Organization | Alibaba | Sarvam |
| Release Date | Jul 22, 2025 | Mar 6, 2026 |
| Performance & Speed | ||
| Throughput | 70.1 tok/s | 89.0 tok/s |
| TTFT | 1646ms | 1218ms |
| Latency | 1646ms | 23680ms |
| Composite Indices | ||
| Value Score | 14.3 | 100.0 |
| Reasoning Score | 44.5 | 34.0 |
| Intelligence | 24.8 | 18.2 |
| Coding | 24.6 | 9.8 |
| Math | 39.3 | — |
| Standard Benchmarks | ||
| GPQA | 61.8% | 73.8% |
| MMLU Pro | 78.8% | — |
| HLE | 4.4% | 10.1% |
| LiveCodeBench | 58.5% | — |
| MATH 500 | 94.2% | — |
| AIME 2025 | 39.3% | — |
| AIME (Original) | 47.7% | — |
| SciCode | 35.9% | 26.4% |
| LCR | 42.3% | 0.0% |
| IFBench | 40.5% | 34.4% |
| TAU-bench v2 | 43.6% | 46.8% |
| TerminalBench Hard | 18.9% | 1.5% |
Key Takeaways
105B (high) offers the best value at $0.04/1M, making it ideal for high-volume applications and cost-conscious projects.
Qwen3 Coder 480B A35B Instruct has the strongest reasoning profile with a 44.5 reasoning score, combining the available reasoning-heavy benchmarks.
Qwen3 Coder 480B A35B Instruct reaches a 24.6 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 Coder 480B A35B Instruct
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
105B (high)
- Cost-sensitive applications
- High-volume processing
