Compare/DeepSeek V3.1 Terminus (Non-reasoning) vs GLM-4.6V (Reasoning)

DeepSeek V3.1 Terminus (Non-reasoning)vsGLM-4.6V (Reasoning)

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

DeepSeek

DeepSeek V3.1 Terminus (Non-reasoning)

Input
$0.27/M
Output
$1/M
Speed
TTFT
Z AI

GLM-4.6V (Reasoning)

Input
$0.3/M
Output
$0.9/M
Speed
29 tok/s
TTFT
1.54s

Winner by Category

Cheaper
GLM-4.6V (Reasoning)
Faster (tok/s)
GLM-4.6V (Reasoning)
Lower Latency
DeepSeek V3.1 Terminus (Non-reasoning)
Benchmarks (8-3)
DeepSeek V3.1 Terminus (Non-reasoning)

Pricing Comparison

MetricDeepSeek V3.1 Terminus (Non-reasoning)GLM-4.6V (Reasoning)
Input ($/M tokens)$0.27$0.3
Output ($/M tokens)$1$0.9
Cost for 1M input + 100K output tokens:
DeepSeek V3.1 Terminus (Non-reasoning)$0.37
GLM-4.6V (Reasoning)$0.39

Speed Comparison

Output Speed (tokens/s) — higher is better
DeepSeek V3.1 Terminus (Non-reasoning)
GLM-4.6V (Reasoning)
29 tok/s
Time to First Token (seconds) — lower is better
DeepSeek V3.1 Terminus (Non-reasoning)
GLM-4.6V (Reasoning)
1.54s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
28.523.4
Coding Index
31.919.7
Math Index
53.785.3
GPQA Diamond
75.1%71.9%
MMLU-Pro
83.6%79.9%
LiveCodeBench
52.9%16.0%
AIME 2025
53.7%85.3%
MATH-500
Humanity's Last Exam
8.4%8.9%
SciCode
32.1%30.4%
IFBench
41.2%30.1%
TerminalBench
31.8%14.4%
DeepSeek V3.1 Terminus (Non-reasoning)8 wins
3 winsGLM-4.6V (Reasoning)

Frequently Asked Questions

Which is cheaper, DeepSeek V3.1 Terminus (Non-reasoning) or GLM-4.6V (Reasoning)?

GLM-4.6V (Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.45/M tokens vs $0.45/M for DeepSeek V3.1 Terminus (Non-reasoning).

Which model performs better on benchmarks?

DeepSeek V3.1 Terminus (Non-reasoning) wins 8 out of 12 benchmarks compared to 3 for GLM-4.6V (Reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GLM-4.6V (Reasoning) generates tokens faster at 29 tok/s vs 0 tok/s. DeepSeek V3.1 Terminus (Non-reasoning) also has lower time-to-first-token (0.00s vs 1.54s).

When should I use DeepSeek V3.1 Terminus (Non-reasoning) vs GLM-4.6V (Reasoning)?

Choose based on your priorities: GLM-4.6V (Reasoning) for lower cost, DeepSeek V3.1 Terminus (Non-reasoning) for stronger benchmark performance, and GLM-4.6V (Reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.