GPT-5.3 Codex (xhigh): benchmark scores, price and ranking

GPT-5.3 Codex (xhigh) from OpenAI ranks #24 of 743 LLMs on the LLMs Tiger aggregate benchmark score (54.3), based on 2 benchmark results captured up to 2026-10-13.

Organisation: OpenAI
Type: Frontier model
Open weights: No
Released: 2026-02-05 (source: epoch)
Cheapest API price: $1.75 in / $14.00 out per million tokens (via aihubmix, observed 2026-10-13T20:47Z)
Aggregate rank: #24 of 743 (score 54.3)

Across the 2 benchmarks where at least ten models are scored, GPT-5.3 Codex (xhigh) is in the top 10% on 2 and in the top half on 2.

Its cheapest listed input price, $1.75 per million tokens, is lower than that of 24% of the 427 models with a listed price.

GPT-5.3 Codex (xhigh) benchmark results

GPT-5.3 Codex (xhigh) benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
WeirdML	77.9%	#9 of 109	Epoch AI Benchmarking Hub	2026-10-13
ALE-Bench	1,655	#7 of 79	Epoch AI Benchmarking Hub	2026-10-13

About GPT-5.3 Codex (xhigh)

GPT-5.3 Codex (xhigh) is a frontier model from OpenAI. It was released on 2026-02-05, according to Epoch AI. The cheapest API listing we track charges $1.75 per million input tokens and $14.00 per million output tokens, through aihubmix. We hold 2 benchmark results for it across 2 areas, and 2 are in the top tenth of their boards.

Where GPT-5.3 Codex (xhigh) stands

On our aggregate score, GPT-5.3 Codex (xhigh) is #24 of 743 ranked models, in the top fifty. Its score is 54.3; the scale is centred on 50, so it sits 4.3 points above the typical model. That beats 97% of the ranked field. Among the 88 ranked OpenAI models it is #9. The next one up is GPT-5.6 Luna (max) at 55.1. The next one down is GPT-6 Luna (max) at 53.7.

GPT-5.3 Codex (xhigh), benchmark by benchmark

On WeirdML, GPT-5.3 Codex (xhigh) scores 77.9% and ranks #9 of 109. Only 8 models score higher, led by GPT-6 Astra (max) at 93.3%, which is 15.4 points ahead. It covers unusual machine-learning tasks. The board median is 41.2%, so GPT-5.3 Codex (xhigh) is 36.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

On ALE-Bench, GPT-5.3 Codex (xhigh) scores 1,655 and ranks #7 of 79. Only 6 models score higher, led by GPT-6 Astra (max) at 2,951, which is 1,296 points ahead. It covers AtCoder heuristic-optimisation contests (performance rating). The board median is 764, so GPT-5.3 Codex (xhigh) is 891 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Strengths and weaknesses

GPT-5.3 Codex (xhigh) 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-5.3 Codex (xhigh) costs more than the typical listing. Models that score within 1.5 points or better and cost less: DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), GLM-5.3 (max) ($0.079, 53.8), GPT-6 Luna (max) ($0.100, 53.7). Similar scores cost more on gemini-3.1-pro-preview-high ($2.00) and Gemini 3 Pro Preview ($2.00).

Release history

The OpenAI model released before it was GPT-5.2 Codex on 2025-12-18, which scores 52.2. The next one was gpt-5.2-chat-latest-20260210 on 2026-02-10, at 49.1.

Reading these numbers

The 2 scores on this page come from 1 source: Epoch AI Benchmarking Hub (2). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 2 other reasoning-effort settings of this model, which redirect here. This page shows the highest. With 2 results the aggregate rank can move a lot when one new score arrives.

GPT-5.3 Codex (xhigh) is also listed at 2 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.