Compare/LFM2.5-1.2B-Instruct vs GPT-5.4 mini (xhigh)

LFM2.5-1.2B-InstructvsGPT-5.4 mini (xhigh)

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

Liquid AI

LFM2.5-1.2B-Instruct

Input
$0/M
Output
$0/M
Speed
TTFT
OpenAI

GPT-5.4 mini (xhigh)

Input
$0.75/M
Output
$4.5/M
Speed
219 tok/s
TTFT
3.82s

Winner by Category

Cheaper
LFM2.5-1.2B-Instruct
Faster (tok/s)
GPT-5.4 mini (xhigh)
Lower Latency
LFM2.5-1.2B-Instruct
Benchmarks (0-7)
GPT-5.4 mini (xhigh)

Pricing Comparison

MetricLFM2.5-1.2B-InstructGPT-5.4 mini (xhigh)
Input ($/M tokens)$0$0.75
Output ($/M tokens)$0$4.5
Cost for 1M input + 100K output tokens:
LFM2.5-1.2B-Instruct$0.00
GPT-5.4 mini (xhigh)$1.20

Speed Comparison

Output Speed (tokens/s) — higher is better
LFM2.5-1.2B-Instruct
GPT-5.4 mini (xhigh)
219 tok/s
Time to First Token (seconds) — lower is better
LFM2.5-1.2B-Instruct
GPT-5.4 mini (xhigh)
3.82s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
8.048.1
Coding Index
0.851.5
Math Index
GPQA Diamond
32.6%87.5%
MMLU-Pro
LiveCodeBench
AIME 2025
MATH-500
Humanity's Last Exam
6.8%26.6%
SciCode
2.3%49.9%
IFBench
43.8%73.3%
TerminalBench
0.0%52.3%
LFM2.5-1.2B-Instruct0 wins
7 winsGPT-5.4 mini (xhigh)

Frequently Asked Questions

Which is cheaper, LFM2.5-1.2B-Instruct or GPT-5.4 mini (xhigh)?

LFM2.5-1.2B-Instruct is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $1.69/M for GPT-5.4 mini (xhigh).

Which model performs better on benchmarks?

GPT-5.4 mini (xhigh) wins 7 out of 12 benchmarks compared to 0 for LFM2.5-1.2B-Instruct. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GPT-5.4 mini (xhigh) generates tokens faster at 219 tok/s vs 0 tok/s. LFM2.5-1.2B-Instruct also has lower time-to-first-token (0.00s vs 3.82s).

When should I use LFM2.5-1.2B-Instruct vs GPT-5.4 mini (xhigh)?

Choose based on your priorities: LFM2.5-1.2B-Instruct for lower cost, GPT-5.4 mini (xhigh) for stronger benchmark performance, and GPT-5.4 mini (xhigh) for faster generation. For latency-sensitive apps, check the TTFT comparison above.