# GPT-5-Codex vs Qwen-14B

> GPT-5-Codex is the stronger model overall, scoring 37.9 to 31.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-codex-vs-qwen-14b
- Last updated: 2026-10-11
- Shared benchmarks: 0

## Summary

- The widest gap is in reasoning, where GPT-5-Codex leads 30.9 to 19.6.
- Qwen-14B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5-Codex | Qwen-14B |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 37.9 | 31.4 |
| Rank | 192 | 275 |
| Context | 400K | — |
| Input $/M | $1.25 | — |
| Output $/M | $10 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-5-Codex: 42.4 (#103)
- Qwen-14B: 31.2 (#288)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| WeirdML | 54.5% | — |
| LMArena Coding | — | 1071 |

## Agentic & Tool Use

- GPT-5-Codex: 31.0 (#72)
- Qwen-14B: —

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| Terminal-Bench | 44.3% | — |

## Reasoning

- GPT-5-Codex: 30.9 (#83)
- Qwen-14B: 19.6 (#257)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | — |
| LMArena Hard Prompts | — | 1027 |
| BIG-Bench Hard | — | 55% |
| Epoch Capabilities Index | — | 113.03 |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |

## Math

- GPT-5-Codex: —
- Qwen-14B: 31.2 (#227)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| LMArena Math | — | 1068 |
| GSM8K | — | 61.3% |

## Knowledge

- GPT-5-Codex: —
- Qwen-14B: —

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
| MMLU | — | 66.3% |

## Multilingual

- GPT-5-Codex: —
- Qwen-14B: 27.5 (#275)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| LMArena Non-English | — | 1041 |
| LMArena Chinese | — | 1077 |

## Instruction Following

- GPT-5-Codex: —
- Qwen-14B: 52.4 (#289)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | — | 1031 |

## Long Context

- GPT-5-Codex: —
- Qwen-14B: 31.3 (#280)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| LMArena Longer Query | — | 1028 |

## Writing & Preference

- GPT-5-Codex: —
- Qwen-14B: 27.6 (#299)

| Benchmark | GPT-5-Codex | Qwen-14B |
|---|---|---|
| LMArena Text | — | 1051 |
| LMArena Creative Writing | — | 1028 |
| LMArena Multi-Turn | — | 1022 |

## FAQ

### Is GPT-5-Codex better than Qwen-14B?

GPT-5-Codex is the stronger model overall, scoring 37.9 to 31.4 on the Noometry Index.

### Is GPT-5-Codex or Qwen-14B better for coding?

GPT-5-Codex scores higher on coding benchmarks: 42.4 versus 31.2 in the Noometry coding category.

### How many benchmarks do GPT-5-Codex and Qwen-14B share?

0 benchmarks have published results for both models. GPT-5-Codex has 3 scored results on Noometry and Qwen-14B has 18.