Model Comparison
Hermes 4 - Llama-3.1 405B (Non-reasoning)
vs. Medium 3
Comparing 2 AI models · 12 benchmarks · Nous Research, Mistral
Recommended Pick
Mi Medium 3 16 metric wins Strongest on: Value, Input price, Output price
Best Value
Medium 3
100.0 value score
39.5 reasoning / $0.80/1M
Lowest Price
Medium 3
$0.40/1M input price
Best Reasoning
Medium 3
39.5 reasoning score
Blends available reasoning benchmarks
Best for Coding
Hermes 4 - Llama-3.1 405B (Non-reasoning)
18.1 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
Medium 3 has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Medium 3 is 2.5x cheaper on input tokens than Hermes 4 - Llama-3.1 405B (Non-reasoning).
Speed gap
Medium 3 generates about 1.2x as many tokens per second as Hermes 4 - Llama-3.1 405B (Non-reasoning).
Reasoning gap
Medium 3 leads Hermes 4 - Llama-3.1 405B (Non-reasoning) by 18.3 points on reasoning.
Coding gap
Hermes 4 - Llama-3.1 405B (Non-reasoning) leads Medium 3 by 4.5 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Hermes 4 - Llama-3.1 405B (Non-reasoning)
Nous Research
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Medium 3
Mistral
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | No Hermes 4 - Llama-3.1 405B (Non-reasoning) | Top Pick Mi Medium 3 |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $1.00/1M | $0.40/1M |
| Output Cost | $3.00/1M | $2.00/1M |
| Blended (3:1) | $1.50/1M | $0.80/1M |
| Specifications | ||
| Organization | Nous Research | Mistral |
| Release Date | Aug 27, 2025 | May 7, 2025 |
| Performance & Speed | ||
| Throughput | 39.1 tok/s | 45.0 tok/s |
| TTFT | 752ms | 494ms |
| Latency | 752ms | 494ms |
| Composite Indices | ||
| Value Score | 28.7 | 100.0 |
| Reasoning Score | 21.2 | 39.5 |
| Intelligence | 17.6 | 18.8 |
| Coding | 18.1 | 13.6 |
| Math | 15.3 | 30.3 |
| Standard Benchmarks | ||
| GPQA | 53.6% | 57.8% |
| MMLU Pro | 72.9% | 76.0% |
| HLE | 4.2% | 4.3% |
| LiveCodeBench | 54.6% | 40.0% |
| MATH 500 | — | 90.7% |
| AIME 2025 | 15.3% | 30.3% |
| AIME (Original) | — | 44.0% |
| SciCode | 34.6% | 33.1% |
| LCR | 20.0% | 28.0% |
| IFBench | 34.8% | 39.3% |
| TAU-bench v2 | 26.6% | 24.3% |
| TerminalBench Hard | 9.8% | 3.8% |
Key Takeaways
Medium 3 offers the best value at $0.40/1M, making it ideal for high-volume applications and cost-conscious projects.
Medium 3 has the strongest reasoning profile with a 39.5 reasoning score, combining the available reasoning-heavy benchmarks.
Hermes 4 - Llama-3.1 405B (Non-reasoning) reaches a 18.1 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
Hermes 4 - Llama-3.1 405B (Non-reasoning)
- Code generation
- Software development
Medium 3
- Cost-sensitive applications
- High-volume processing
- Complex reasoning tasks
- Research & analysis
