Model Comparison
gpt-oss-120b (high)
vs. M2.5
Comparing 2 AI models · 10 benchmarks · OpenAI, MiniMax
Recommended Pick
Mi M2.5 9 metric wins Strongest on: Intelligence, Coding, GPQA
Best Value
gpt-oss-120b (high)
100.0 value score
63.4 reasoning / $0.26/1M
Lowest Price
gpt-oss-120b (high)
$0.15/1M input price
Best Reasoning
gpt-oss-120b (high)
63.4 reasoning score
Blends available reasoning benchmarks
Best for Coding
M2.5
37.4 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
gpt-oss-120b (high) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
gpt-oss-120b (high) is 2.0x cheaper on input tokens than M2.5.
Speed gap
gpt-oss-120b (high) generates about 1.6x as many tokens per second as M2.5.
Reasoning gap
gpt-oss-120b (high) leads M2.5 by 14.8 points on reasoning.
Coding gap
M2.5 leads gpt-oss-120b (high) by 8.8 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
gpt-oss-120b (high)
OpenAI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
M2.5
MiniMax
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Op gpt-oss-120b (high) | Top Pick Mi M2.5 |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $0.15/1M | $0.30/1M |
| Output Cost | $0.60/1M | $1.20/1M |
| Blended (3:1) | $0.26/1M | $0.52/1M |
| Specifications | ||
| Organization | OpenAI | MiniMax |
| Release Date | Aug 5, 2025 | Feb 12, 2026 |
| Performance & Speed | ||
| Throughput | 353.8 tok/s | 217.6 tok/s |
| TTFT | 517ms | 12625ms |
| Latency | 6171ms | 21815ms |
| Composite Indices | ||
| Value Score | 100.0 | 38.3 |
| Reasoning Score | 63.4 | 48.6 |
| Intelligence | 33.3 | 41.9 |
| Coding | 28.6 | 37.4 |
| Math | 93.4 | — |
| Standard Benchmarks | ||
| GPQA | 78.2% | 84.8% |
| MMLU Pro | 80.8% | — |
| HLE | 18.5% | 19.1% |
| LiveCodeBench | 87.8% | — |
| AIME 2025 | 93.4% | — |
| SciCode | 38.9% | 42.6% |
| LCR | 50.7% | 66.0% |
| IFBench | 69.0% | 71.6% |
| TAU-bench v2 | 65.8% | 95.3% |
| TerminalBench Hard | 23.5% | 34.8% |
Key Takeaways
gpt-oss-120b (high) offers the best value at $0.15/1M, making it ideal for high-volume applications and cost-conscious projects.
gpt-oss-120b (high) has the strongest reasoning profile with a 63.4 reasoning score, combining the available reasoning-heavy benchmarks.
M2.5 reaches a 37.4 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
gpt-oss-120b (high)
- Cost-sensitive applications
- High-volume processing
- Complex reasoning tasks
- Research & analysis
M2.5
- Code generation
- Software development
