Compare/GLM-4.7-Flash (Non-reasoning) vs Hermes 4 - Llama-3.1 70B (Reasoning)

GLM-4.7-Flash (Non-reasoning)vsHermes 4 - Llama-3.1 70B (Reasoning)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Z AI

GLM-4.7-Flash (Non-reasoning)

Input
$0.07/M
Output
$0.4/M
Speed
90 tok/s
TTFT
0.67s
Nous Research

Hermes 4 - Llama-3.1 70B (Reasoning)

Input
$0.13/M
Output
$0.4/M
Speed
81 tok/s
TTFT
0.58s

Winner by Category

Cheaper
GLM-4.7-Flash (Non-reasoning)
Faster (tok/s)
GLM-4.7-Flash (Non-reasoning)
Lower Latency
Hermes 4 - Llama-3.1 70B (Reasoning)
Benchmarks (2-9)
Hermes 4 - Llama-3.1 70B (Reasoning)

Pricing Comparison

MetricGLM-4.7-Flash (Non-reasoning)Hermes 4 - Llama-3.1 70B (Reasoning)
Input ($/M tokens)$0.07$0.13
Output ($/M tokens)$0.4$0.4
Cost for 1M input + 100K output tokens:
GLM-4.7-Flash (Non-reasoning)$0.11
Hermes 4 - Llama-3.1 70B (Reasoning)$0.17

Speed Comparison

Output Speed (tokens/s) — higher is better
GLM-4.7-Flash (Non-reasoning)
90 tok/s
Hermes 4 - Llama-3.1 70B (Reasoning)
81 tok/s
Time to First Token (seconds) — lower is better
GLM-4.7-Flash (Non-reasoning)
0.67s
Hermes 4 - Llama-3.1 70B (Reasoning)
0.58s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
22.116.0
Coding Index
11.014.4
Math Index
68.7
GPQA Diamond
45.2%69.9%
MMLU-Pro
81.1%
LiveCodeBench
65.3%
AIME 2025
68.7%
MATH-500
Humanity's Last Exam
4.9%7.9%
SciCode
25.5%34.1%
IFBench
46.3%31.3%
TerminalBench
3.8%4.5%
GLM-4.7-Flash (Non-reasoning)2 wins
9 winsHermes 4 - Llama-3.1 70B (Reasoning)

Frequently Asked Questions

Which is cheaper, GLM-4.7-Flash (Non-reasoning) or Hermes 4 - Llama-3.1 70B (Reasoning)?

GLM-4.7-Flash (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.15/M tokens vs $0.20/M for Hermes 4 - Llama-3.1 70B (Reasoning).

Which model performs better on benchmarks?

Hermes 4 - Llama-3.1 70B (Reasoning) wins 9 out of 12 benchmarks compared to 2 for GLM-4.7-Flash (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GLM-4.7-Flash (Non-reasoning) generates tokens faster at 90 tok/s vs 81 tok/s. However, Hermes 4 - Llama-3.1 70B (Reasoning) has lower time-to-first-token (0.58s vs 0.67s).

When should I use GLM-4.7-Flash (Non-reasoning) vs Hermes 4 - Llama-3.1 70B (Reasoning)?

Choose based on your priorities: GLM-4.7-Flash (Non-reasoning) for lower cost, Hermes 4 - Llama-3.1 70B (Reasoning) for stronger benchmark performance, and GLM-4.7-Flash (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.