Model comparison
GPT-5.3 Codex vs Qwen3.6-27B
Head-to-head evidence from 16 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: GPT-5.3 Codex #26 (Supported); Qwen3.6-27B #93 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GPT-5.3 Codex and Qwen3.6-27B share 16 comparable benchmark results. 2 of 8 categories are comparable. 5 results are unique to GPT-5.3 Codex; 38 to Qwen3.6-27B.
Updated July 21, 2026- Shared results
- 16
- GPT-5.3 Codex only
- 5
- Qwen3.6-27B only
- 38
- Comparable categories
- 2 / 8
Pick GPT-5.3 Codex if you want the stronger benchmark profile. Qwen3.6-27B only becomes the better choice if coding is the priority or you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 16 shared benchmark results across 6 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
GPT-5.3 Codex is clearly ahead on the BenchAlign aggregate, 66.69 to 53.82. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.3 Codex's sharpest advantage is in agentic, where it averages 71.4 against 59.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 77.3% to 59.3%. Qwen3.6-27B does hit back in coding, so the answer changes if that is the part of the workload you care about most.
GPT-5.3 Codex is also the more expensive model on tokens at $1.75 input / $14.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.6-27B. That is roughly Infinityx on output cost alone. GPT-5.3 Codex gives you the larger context window at 400K, compared with 262K for Qwen3.6-27B.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | GPT-5.3 Codex | Δ | Qwen3.6-27B |
|---|---|---|---|
| Agentic | GPT-5.3 Codex71.4 | Margin← 12.1 | Qwen3.6-27B59.3 |
| Coding | GPT-5.3 Codex67.2 | Margin→ 10.3 | Qwen3.6-27B77.5 |
| Knowledge | GPT-5.3 CodexNot measured | MarginNo overlap | Qwen3.6-27B53.3 |
| Math | GPT-5.3 CodexNot measured | MarginNo overlap | Qwen3.6-27B89.2 |
| Multimodal | GPT-5.3 CodexNot measured | MarginNo overlap | Qwen3.6-27B76.7 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 77.3%B 59.3%Winner: GPT-5.3 CodexΔ 18Terminal-Bench 2.0: GPT-5.3 Codex scored 77.3%; Qwen3.6-27B scored 59.3%. GPT-5.3 Codex wins this benchmark. - Source ↗
SWE-bench Verified
CodingA 85%B 77.2%Winner: GPT-5.3 CodexΔ 7.8SWE-bench Verified: GPT-5.3 Codex scored 85%; Qwen3.6-27B scored 77.2%. GPT-5.3 Codex wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 56.8%B 53.5%Winner: GPT-5.3 CodexΔ 3.3SWE-bench Pro: GPT-5.3 Codex scored 56.8%; Qwen3.6-27B scored 53.5%. GPT-5.3 Codex wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | GPT-5.3 Codex | Qwen3.6-27B | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GPT-5.3 Codex$1.75 input / $14 output | Qwen3.6-27B$0 input / $0 output | Qwen3.6-27B has the lower combined listed price. |
| Generation speedtokens per second | GPT-5.3 Codex79 tok/s | Qwen3.6-27BNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | GPT-5.3 Codex88.26 s | Qwen3.6-27BNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | GPT-5.3 Codex400K | Qwen3.6-27B262K | GPT-5.3 Codex lists the larger context window. |
Benchmark Deep Dive
AgenticGPT-5.3 Codex wins12 benchmarks
| Benchmark | GPT-5.3 Codex | Qwen3.6-27B | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 77.3% | 59.3% | GPT-5.3 Codex leads |
| OSWorld-VerifiedSource | 64.7% | — | Not comparable |
| τ²-bench resultsSource | 86% | 94.2% | Qwen3.6-27B leads |
| Gert LabsSource | 57.47% | 54.84% | GPT-5.3 Codex leads |
| JobBenchSource | 33.7% | — | Not comparable |
| Claw-EvalSource | — | 72.4% | Not comparable |
| QwenClawBenchSource | — | 53.4% | Not comparable |
| QwenWebBenchSource | — | 1487 | Not comparable |
| AndroidWorldSource | — | 70.3% | Not comparable |
| AA Agentic IndexSource | — | 27.0% | Not comparable |
| GDPval-AASource | — | 32.0% | Not comparable |
| GDPval-AASource | — | 1140 | Not comparable |
CodingQwen3.6-27B wins10 benchmarks
| Benchmark | GPT-5.3 Codex | Qwen3.6-27B | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 85% | 77.2% | GPT-5.3 Codex leads |
| SWE-bench ProSource | 56.8% | 53.5% | GPT-5.3 Codex leads |
| SWE-RebenchSource | 58.2% | — | Not comparable |
| Vibe Code BenchSource | 61.77% | — | Not comparable |
| AA-SciCodeSource | 53.2% | 39.8% | GPT-5.3 Codex leads |
| SWE MultilingualSource | — | 71.3% | Not comparable |
| Terminal-Bench 2.0Source | — | 59.3% | Not comparable |
| LiveCodeBenchSource | — | 83.9% | Not comparable |
| NL2RepoSource | — | 36.2% | Not comparable |
| AA Coding IndexSource | — | 53.7% | Not comparable |
Reasoning2 benchmarks
Knowledge12 benchmarks
| Benchmark | GPT-5.3 Codex | Qwen3.6-27B | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 44.3% | 37.0% | GPT-5.3 Codex leads |
| AA-GPQA DiamondSource | 91.5% | 84.2% | GPT-5.3 Codex leads |
| AA-HLESource | 39.9% | 21.6% | GPT-5.3 Codex leads |
| AA-Omniscience IndexSource | 9.9% | -19.8% | GPT-5.3 Codex leads |
| AA-Omniscience AccuracySource | 51.8% | 19.2% | GPT-5.3 Codex leads |
| AA-Omniscience Hallucination RateSource | 86.9% | 48.3% | Qwen3.6-27B leads |
| MMLU-ProSource | — | 86.2% | Not comparable |
| MMLU-ReduxSource | — | 93.5% | Not comparable |
| SuperGPQASource | — | 66% | Not comparable |
| C-EvalSource | — | 91.4% | Not comparable |
| GPQASource | — | 87.8% | Not comparable |
| HLESource | — | 24% | Not comparable |
Math5 benchmarks
Multimodal17 benchmarks
| Benchmark | GPT-5.3 Codex | Qwen3.6-27B | Result |
|---|---|---|---|
| AA-MMMU-ProSource | 78.5% | 74.6% | GPT-5.3 Codex leads |
| Design Arena WebsiteSource | 1193 | — | Not comparable |
| MMMUSource | — | 82.9% | Not comparable |
| MMMU-ProSource | — | 75.8% | Not comparable |
| RealWorldQASource | — | 84.1% | Not comparable |
| DynaMathSource | — | 85.6% | Not comparable |
| MStarSource | — | 81.4% | Not comparable |
| SimpleVQASource | — | 56.1% | Not comparable |
| CharXivSource | — | 78.4% | Not comparable |
| CC-OCRSource | — | 81.2% | Not comparable |
| CountBenchSource | — | 97.8% | Not comparable |
| RefCOCO (avg)Source | — | 92.5% | Not comparable |
| ERQASource | — | 62.5% | Not comparable |
| Video-MME (with subtitle)Source | — | 87.7% | Not comparable |
| VideoMMMUSource | — | 84.4% | Not comparable |
| MLVU (M-Avg)Source | — | 86.6% | Not comparable |
| V*Source | — | 94.7% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | GPT-5.3 Codex | Qwen3.6-27B | Result |
|---|---|---|---|
| AA-IFBenchSource | 75.4% | 67.6% | GPT-5.3 Codex leads |
Frequently Asked Questions (3)
Which is better, GPT-5.3 Codex or Qwen3.6-27B?
GPT-5.3 Codex is ahead on BenchLM's BenchAlign leaderboard, 66.69 to 53.82. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 77.3% and 59.3%.
Which is better for coding, GPT-5.3 Codex or Qwen3.6-27B?
Qwen3.6-27B has the edge for coding in this comparison, averaging 77.5 versus 67.2. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, GPT-5.3 Codex or Qwen3.6-27B?
GPT-5.3 Codex has the edge for agentic tasks in this comparison, averaging 71.4 versus 59.3. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.