Skip to main content

GPT-5.4 mini vs GPT-6.1 Sol

SpecGPT-5.4 miniGPT-6.1 Sol
Respan Index34.1 (#82)69.7 (#12)
Blended price per 1M$1.69$4
Context window400K1.1M
Input / output per 1M$0.75 / $4.5$2 / $10
Max output128K128K
Inputtext, visiontext, vision
Released2026-03-172026-09-29

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryGPT-5.4 miniGPT-6.1 Sol
Coding13.1 GPT-5.4 mini52.7 GPT-6.1 Sol
Agents48.5 GPT-5.4 mini72.9 GPT-6.1 Sol
Reasoning32.6 GPT-5.4 mini87.6 GPT-6.1 Sol
Knowledge & language28.9 GPT-5.4 mini74.4 GPT-6.1 Sol
Human preference51.1 GPT-5.4 mini59.7 GPT-6.1 Sol

How the Respan Index works

Benchmarks both report

GPT-6.1 Sol leads on 18 of the 19 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkGPT-5.4 miniGPT-6.1 Sol
LMArena WebDev (LMArena)13981743
LMArena Elo14471483
LMArena Vision (LMArena)12501290
Chess Puzzles (Epoch AI)23%68%
FrontierMath Tier 4 v2 (Epoch AI)9.8%100%
FrontierMath Tiers 1-3 v2 (Epoch AI)51.2%93.7%
GPQA Diamond88%95.4%
LiveBench Agentic Coding (LiveBench)41.7%56.8%
LiveBench Coding (LiveBench)71.6%80.7%
LiveBench Data Analysis (LiveBench)70.8%82.7%
LiveBench Instruction Following (LiveBench)59.8%74.2%
LiveBench Language (LiveBench)71%90.1%
LiveBench Mathematics (LiveBench)78.5%96.8%
LiveBench Reasoning (LiveBench)71.3%92.6%
Mystery Game Puzzles (Epoch AI)11%80%
OTIS Mock AIME 2024-2025 (Epoch AI)88.9%100%
SAGE (Vals AI)50.8%46.5%
SimpleQA Verified29.4%73.9%
Terminal-Bench 4.0 (Vals AI)2.5%55.1%

FAQ

Is GPT-5.4 mini better than GPT-6.1 Sol?

GPT-6.1 Sol scores 69.7 on the Respan Index (#12), ahead of GPT-5.4 mini at 34.1 (#82). GPT-6.1 Sol leads on 18 of the 19 benchmarks both report.

Which is cheaper, GPT-5.4 mini or GPT-6.1 Sol?

GPT-5.4 mini lists at $0.75 input and $4.5 output per 1M tokens; GPT-6.1 Sol at $2 and $10. GPT-5.4 mini is 2.4x cheaper per token at list price.

Which has the larger context window?

GPT-5.4 mini accepts 400K tokens and GPT-6.1 Sol 1.1M tokens.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.