Hermes 4 – Llama-3.1 70B (Reasoning)
Intelligence
16
#317/523
Coding
14.4
#269/429
Math
68.7
#98/265
Speed
81 tok/s
TTFT: 578.00s
Pricing
$0.13 / $0.40
per 1M tokens (in/out)
Hermes 4 – Llama-3.1 70B (Reasoning) is Nous Research’s advanced AI model designed for complex reasoning tasks. It processes at 81.017 tokens per second and is priced at $0.13 per million input tokens, targeting professional users in fields requiring analytical capabilities.
When to Use Hermes 4 – Llama-3.1 70B (Reasoning)
✓ Best For
- Complex reasoning tasks
- Data analysis and interpretation
- Programming assistance
✗ Not Ideal For
- Basic conversational applications
- High-speed real-time interactions
How Hermes 4 – Llama-3.1 70B (Reasoning) Compares
Intelligence Index · Higher is better
NanbeigeMistralNous ResearchGoogleDeepSeek
Benchmark Profile
Coding Index
BaiduOpenAINous ResearchAlibaba
Output Speed · tok/s
OpenAINous ResearchMistralMeta
Math Index
NVIDIAGoogleNous ResearchAlibabaDeepSeek
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 16 | |
| Coding Index | 14.4 | |
| Math Index | 68.7 | |
| MMLU-Pro | 811% | |
| GPQA | 699% | |
| LiveCodeBench | 653% | |
| HLE | 79% | |
| SciCode | 34.1% | |
| IFBench | 31.3% | |
| LCR | 6.7% | |
| TerminalBench Hard | 4.5% | |
| Tau2 | 22.5% | |
| AIME 2025 | 68.7% |
Data: Artificial Analysis · Updated: March 26, 2026
Frequently Asked Questions (15)
When was Hermes 4 - Llama-3.1 70B (Reasoning) released?
Hermes 4 - Llama-3.1 70B (Reasoning) was released on August 27, 2025.
Who created Hermes 4 - Llama-3.1 70B (Reasoning)?
Hermes 4 - Llama-3.1 70B (Reasoning) was created by Nous Research.
How intelligent is Hermes 4 - Llama-3.1 70B (Reasoning)?
Hermes 4 - Llama-3.1 70B (Reasoning) scores 16 on the Artificial Analysis Intelligence Index, placing it above average among other open weight models of similar size (median: 15).
How fast is Hermes 4 - Llama-3.1 70B (Reasoning)?
Hermes 4 - Llama-3.1 70B (Reasoning) generates output at 72.4 tokens per second (based on the median across providers serving the model), which is below average compared to other open weight models of similar size (median: 76.2 t/s).
What is the latency of Hermes 4 - Llama-3.1 70B (Reasoning)?
Hermes 4 - Llama-3.1 70B (Reasoning) has a time to first token (TTFT) of 1.41s (based on the median across providers serving the model), which is better than average compared to other open weight models of similar size (median: 1.55s).
How much does Hermes 4 - Llama-3.1 70B (Reasoning) cost?
Hermes 4 - Llama-3.1 70B (Reasoning) costs $0.13 per 1M input tokens (very competitive, median: $0.35) and $0.40 per 1M output tokens (very competitive, median: $0.75), based on the median across providers serving the model.
What is Hermes 4 - Llama-3.1 70B (Reasoning) API pricing?
Hermes 4 - Llama-3.1 70B (Reasoning) costs $0.13 per 1M input tokens and $0.40 per 1M output tokens (based on the median across providers serving the model). For a blended rate (3:1 input to output ratio), this is $0.20 per 1M tokens. Pricing may vary by provider.
How verbose is Hermes 4 - Llama-3.1 70B (Reasoning)?
When evaluated on the Intelligence Index, Hermes 4 - Llama-3.1 70B (Reasoning) generated 49M output tokens, which is somewhat higher than average compared to other open weight models of similar size (median: 7.3M).
Is Hermes 4 - Llama-3.1 70B (Reasoning) a reasoning model?
Yes, Hermes 4 - Llama-3.1 70B (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does Hermes 4 - Llama-3.1 70B (Reasoning) support?
Hermes 4 - Llama-3.1 70B (Reasoning) supports text input.
What output modalities does Hermes 4 - Llama-3.1 70B (Reasoning) support?
Hermes 4 - Llama-3.1 70B (Reasoning) supports text output.
Can Hermes 4 - Llama-3.1 70B (Reasoning) process images?
No, Hermes 4 - Llama-3.1 70B (Reasoning) does not support image input. It can only process text.
Is Hermes 4 - Llama-3.1 70B (Reasoning) multimodal?
No, Hermes 4 - Llama-3.1 70B (Reasoning) is not multimodal. It only supports text input.
What is the context window of Hermes 4 - Llama-3.1 70B (Reasoning)?
Hermes 4 - Llama-3.1 70B (Reasoning) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Is Hermes 4 - Llama-3.1 70B (Reasoning) open source?
Yes, Hermes 4 - Llama-3.1 70B (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
