gpt-4-0125-preview: benchmark scores, price and ranking

gpt-4-0125-preview from OpenAI ranks #297 of 743 LLMs on the LLMs Tiger aggregate benchmark score (49.1), based on 2 benchmark results captured up to 2026-10-13.

Organisation: OpenAI
Type: Preview release
Open weights: No
Cheapest API price: $10.00 in / $30.00 out per million tokens (via azure, observed 2026-10-13T21:47Z)
Aggregate rank: #297 of 743 (score 49.1)

Across the 2 benchmarks where at least ten models are scored, gpt-4-0125-preview is in the top 10% on 0 and in the top half on 1.

Its cheapest listed input price, $10.00 per million tokens, is lower than that of 4% of the 427 models with a listed price.

gpt-4-0125-preview benchmark results

gpt-4-0125-preview benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
Aider Code Editing	66.2%	#22 of 92	Aider Aider Code Editing leaderboard	2026-10-13
Aider Refactoring	33.7%	#10 of 12	Aider Aider Refactoring leaderboard	2026-10-13

About gpt-4-0125-preview

gpt-4-0125-preview is a preview release from OpenAI. The cheapest API listing we track charges $10.00 per million input tokens and $30.00 per million output tokens, through azure. We hold 2 benchmark results for it across 2 areas, and none is in the top tenth of its board.

Where gpt-4-0125-preview stands

On our aggregate score, gpt-4-0125-preview is #297 of 743 ranked models, around the middle of the table. Its score is 49.1; the scale is centred on 50, so it sits 0.9 points below the typical model. That beats 60% of the ranked field. Among the 88 ranked OpenAI models it is #41. The next one up is gpt-5-chat at 49.2. The next one down is gpt-5.2-chat-latest-20260210 at 49.1.

gpt-4-0125-preview, benchmark by benchmark

Aider Code Editing puts gpt-4-0125-preview at 66.2%, #22 of 92. That beats 76% of the models tested, but 18.0 points short of Claude 3.5-Sonnet-20241022 (84.2%). The results either side are gemini-1.5-pro-exp-0827 (66.9%) and gpt-4-0314 (66.2%). It covers Aider official benchmark. The board median is 55.6%, so gpt-4-0125-preview is 10.6 points above it. Source: Aider Aider Code Editing leaderboard, captured 2026-10-13.

Aider Refactoring is a weak spot. gpt-4-0125-preview scores 33.7%, #10 of 12, behind 75% of the models measured. The leader, Claude 3.5-Sonnet-20241022, reaches 92.1%. It covers Aider official benchmark. The board median is 50.0%, so gpt-4-0125-preview is 16.3 points below it. Source: Aider Aider Refactoring leaderboard, captured 2026-10-13.

Strengths and weaknesses

gpt-4-0125-preview has too few results on boards with ten or more models to say where it is strong or weak. Treat any single score as a data point, not a verdict.

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so gpt-4-0125-preview costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on o1-preview ($15.00) and Claude Opus 4.1 (unknown thinking) ($15.00).

Reading these numbers

The 2 scores on this page come from 2 sources: Aider Aider Code Editing leaderboard (1), Aider Aider Refactoring leaderboard (1). 2 of them come from an official leaderboard and 0 from datasets or third-party runs. With 2 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#294 kimi-k2-0711-preview (49.1)
#295 TTS(Bo8) (49.1)
#296 DevStral Small 2505 (49.1)
#298 gpt-5.2-chat-latest-20260210 (49.1)
#299 parasail-qwen3-235b-a22b-instruct-2507 (49.0)
#300 Grok-2-mini (49.0)