Model comparison
GPT-5.3 Codex vs Qwen3-Next 80B-A3B Instruct
GPT-5.3 Codex is the stronger model overall, scoring 45.5 to 42.5 on the Noometry Index. Qwen3-Next 80B-A3B Instruct costs 5.5× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Alibaba (Qwen)Summary
- The widest gap is in coding, where GPT-5.3 Codex leads 48.5 to 41.9.
- Qwen3-Next 80B-A3B Instruct is cheaper at $0.50 / $2 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.
- GPT-5.3 Codex accepts more context: 400K tokens versus 131K.
- Qwen3-Next 80B-A3B Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 45.5 | 42.5 |
| Released | 2026-02-05 | 2025-09 |
| Weights | Proprietary | Open |
| Context window | 400K | 131K |
| Max output | 128K | 33K |
| Input $ / M tokens | $1.75 | $0.50 |
| Output $ / M tokens | $14 | $2 |
| Results tracked | 8 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.3 Codex leads
GPT-5.3 Codex: 48.5 (#51), Qwen3-Next 80B-A3B Instruct: 41.9 (#101)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| SWE-bench Verified | 74.8% | — |
| LMArena WebDev | 1408 | — |
| WeirdML | 79.3% | — |
| LMArena Coding | — | 1441 |
| ALE-Bench | 1,655 | — |
Agentic & Tool Use Not comparable
GPT-5.3 Codex: 47.8 (#9), Qwen3-Next 80B-A3B Instruct: —
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Terminal-Bench | 78.4% | — |
| METR Time Horizons | 74.5% | — |
| Vending-Bench 2 | 5,940 | — |
Reasoning Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 30.1 (#79)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Kagi LLM Benchmark | — | 66.7% |
| LMArena Hard Prompts | — | 1428 |
| Epoch Capabilities Index | 156.84 | — |
Math Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 36.3 (#118)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Omni-MATH | — | 46.7% |
| LMArena Math | — | 1440 |
Knowledge Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 41.3 (#102)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| MMLU-Pro | — | 78.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 63% |
| LMArena Expert | — | 1416 |
Multilingual Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 52.3 (#88)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Non-English | — | 1407 |
| LMArena Chinese | — | 1461 |
| LMArena French | — | 1415 |
| LMArena German | — | 1416 |
| LMArena Japanese | — | 1398 |
| LMArena Korean | — | 1366 |
| LMArena Russian | — | 1403 |
| LMArena Spanish | — | 1436 |
Instruction Following Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 71.0 (#154)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| IFEval | — | 81% |
| LMArena Instruction Following | — | 1388 |
Long Context Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 37.0 (#218)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| Fiction.LiveBench | — | 55.6% |
| LMArena Longer Query | — | 1403 |
Writing & Preference Not comparable
GPT-5.3 Codex: —, Qwen3-Next 80B-A3B Instruct: 58.2 (#116)
| Benchmark | GPT-5.3 Codex | Qwen3-Next 80B-A3B Instruct |
|---|---|---|
| LMArena Text | — | 1417 |
| LMArena Creative Writing | — | 1333 |
| WildBench | — | 80.7% |
| LMArena Multi-Turn | — | 1415 |
Frequently asked questions
Is GPT-5.3 Codex better than Qwen3-Next 80B-A3B Instruct?
GPT-5.3 Codex is the stronger model overall, scoring 45.5 to 42.5 on the Noometry Index. Qwen3-Next 80B-A3B Instruct costs 5.5× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.
Which is cheaper, GPT-5.3 Codex or Qwen3-Next 80B-A3B Instruct?
Qwen3-Next 80B-A3B Instruct is cheaper. It lists at $0.50 per million input tokens and $2 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.
Is GPT-5.3 Codex or Qwen3-Next 80B-A3B Instruct better for coding?
GPT-5.3 Codex scores higher on coding benchmarks: 48.5 versus 41.9 in the Noometry coding category.
Which has the bigger context window?
GPT-5.3 Codex does, with 400K tokens against 131K.
How many benchmarks do GPT-5.3 Codex and Qwen3-Next 80B-A3B Instruct share?
0 benchmarks have published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Qwen3-Next 80B-A3B Instruct has 25.