GPT-5.4 minivs.Qwen 3.8 27B
Which one is better?
Level in the overall index: the gap is too small for a clear winner. Qwen 3.8 27B is cheaper: GPT-5.4 mini costs 18.8 times (input) and 1.8 times (output) as much per token.
5
GPT-5.4 mini
Rank 54 to 60
11
Qwen 3.8 27B
Rank 44 to 57
230
GPT-5.4 mini
fast
40
Qwen 3.8 27B
slow
$1.69
GPT-5.4 mini
fairly expensive
$0.66
Qwen 3.8 27B
fairly cheap
GPT-5.4 mini
- Subscription
No OpenAI subscription includes this model. Through other providers from ≈ €6.36 a month (ElevenLabs Starter, with limited credits). 36 plans in Germany.
- Via API
$0.75 input, $4.50 output per 1M tokens, unchanged since our price tracking began on Aug 14, 2026.
Qwen 3.8 27B
- Subscription
No Alibaba subscription includes this model. Through other providers from ≈ €1.06 a month (Command Code Go, with limited credits). 10 plans in Germany.
- Via API
$0.04 input, $2.50 output per 1M tokens according to the provider; the price index has listed it at $0.42 and $3 since Oct 27, 2026.
Overall index with data through Oct 29, 2026 · Prices from Oct 29, 2026 · Speed: measured by Artificial Analysis · Subscriptions as of Oct 24, 2026 · Subscription prices incl. VAT
Where is the difference?
Values from 0 to 100 per task area. We only name a winner when the gap is large enough and there are enough tests.
GPT-5.4 mini ahead: 0Qwen 3.8 27B ahead: 4level: 2too little data: 4of 10 areas
AreaGPT-5.4 miniQwen 3.8 27B0 to 100, higher is betterResult
Areas in the overall index
CodingQwen 3.8 27B aheadGPT-5.4 mini: 5Qwen 3.8 27B: 13
ReasoninglevelGPT-5.4 mini: 7Qwen 3.8 27B: 6
Mathematicsonly Qwen 3.8 27B has a scoreGPT-5.4 mini: not enough dataQwen 3.8 27B: 8
Knowledgetoo little dataGPT-5.4 mini: 4Qwen 3.8 27B: 5
AgentslevelGPT-5.4 mini: 9Qwen 3.8 27B: 21
Generalonly GPT-5.4 mini has a scoreGPT-5.4 mini: 11Qwen 3.8 27B: not enough data
MultimodalQwen 3.8 27B aheadGPT-5.4 mini: 6Qwen 3.8 27B: 13
Specialist fields
FinanceQwen 3.8 27B aheadGPT-5.4 mini: 5Qwen 3.8 27B: 34
LegalQwen 3.8 27B aheadGPT-5.4 mini: 3Qwen 3.8 27B: 23
Healthonly Qwen 3.8 27B has a scoreGPT-5.4 mini: not enough dataQwen 3.8 27B: 3
Show 1 more areas that do not count toward the overall index
AreaGPT-5.4 miniQwen 3.8 27B0 to 100, higher is betterResult
Cybersecurityonly Qwen 3.8 27B has a scoreGPT-5.4 mini: not enough dataQwen 3.8 27B: 19
See for yourself: same task, one attempt
The same prompt for both models. We show the latest attempt without fixes; it does not count toward the overall index.
Pelican on a bicycle
A community classic for free-form SVG drawing.
GPT-5.4 mini
Reasoning (requested): high
Current attempt
- Run date
- Oct 2, 2026
- API provider
- openrouter
- Requested API model
- openai/gpt-5.4-mini
- API model reported in response
- openai/gpt-5.4-mini
- Reasoning setting
- high
- Pinned endpoint
- OpenAI | openai/gpt-5.4-mini-20260317
- Temperature
- Provider default
- Token limit including reasoning
- 8,192 tokens
- Test protocol
- community-visual-tests-v1
- Technical checks
- Passed, not a quality rating
- Response time
- 13.5 s
Visually reviewed. This individual test does not establish overall model quality.
Earlier attempts (2)
- Oct 1, 2026API request failedReasoning setting: high
- Sep 7, 2026API request failedReasoning setting: high
Qwen 3.8 27B
Not tested yet
Prompt and test conditions
Generate an SVG of a pelican riding a bicycle
Token limit including reasoning: 8,192 tokens
Task origin (Simon Willison)SVG short test
SVG short test
One SVG task with the same fixed settings for every model, without hidden thinking.
Shared test round: 43b8c5c46830
GPT-5.4 mini
Unavailable in this round
Qwen 3.8 27B
Test details
API reports 0 reasoning tokensAPI model: qwen/qwen3.8-27b865 output tokensAll data side by side
The better value is bold. For the overall index only when the gap is clear.
| Metric | GPT-5.4 mini | Qwen 3.8 27B |
|---|---|---|
| Intelligence | ||
| Overall index0 to 100, higher is better | 5 | 11 |
| Rank | Rank 54 to 60 of 95 | Rank 44 to 57 of 95 |
| Price per 1M tokens | ||
| Inputwhat you send | $0.75 | $0.04 |
| Outputwhat the model writes | $4.50 | $2.50 |
| Speed | ||
| Tokens per second | 230 | 40 |
| Technical | ||
| Context windowhow much text the model reads at once | 400,000 tokens≈ 600 A4 pages | 262,144 tokens≈ 390 A4 pages |
| Knowledge cutoff | Aug 31, 2025 | n/a |
| Released | Mar 17, 2026 | Aug 14, 2026 |
| Availability | Active | Active |
| Open weights | no | yes |
Prices without caching or batch discounts · Page count as a rule of thumb from Artificial Analysis
Shared benchmarks
Single results from the same benchmark version, task and metric. A value is bold only with the same test setup and a clear gap.
A: GPT-5.4 miniB: Qwen 3.8 27B
Code MigrationOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 12.9%Qwen 3.8 27B: 14.2%Vals AIAs of 10/18/2026
CritPt v2Overall · higher is betterGPT-5.4 mini: 2.9%Qwen 3.8 27B: 0.0%Artificial AnalysisAs of 10/29/2026
EMBOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 45.4%Qwen 3.8 27B: 59.7%Vals AIAs of 10/20/2026
Finance Agent (v2)Overall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 45.4%Qwen 3.8 27B: 48.6%Vals AIAs of 10/20/2026
GDPval-AAOverall · higher is betterGPT-5.4 mini: 597.62 eloQwen 3.8 27B: 1,082.47 eloArtificial AnalysisAs of 10/29/2026
GPQA DiamondOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 83.1%Qwen 3.8 27B: 88.9%Vals AIAs of 09/01/2026
Legal Research BenchOverall · All-pass · higher is betterTest setup not documented as identicalGPT-5.4 mini: 12.5%Qwen 3.8 27B: 36.1%Vals AIAs of 10/26/2026
LiveCodeBenchOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 81.5%Qwen 3.8 27B: 84%Vals AIAs of 09/01/2026
MMLU ProOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 84.6%Qwen 3.8 27B: 84.3%Vals AIAs of 09/01/2026
MMMU ProOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 79.2%Qwen 3.8 27B: 83.9%Vals AIAs of 09/01/2026
Show 54 more benchmarks
MMMU-ProOverall · higher is betterGPT-5.4 mini: 71.2%Qwen 3.8 27B: 74.2%Artificial AnalysisAs of 10/29/2026
MortgageTaxOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 63.5%Qwen 3.8 27B: 64.9%Vals AIAs of 09/01/2026
ProgramBenchOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 0%Qwen 3.8 27B: 0%Vals AIAs of 10/20/2026
SWE-bench VerifiedOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 73%Qwen 3.8 27B: 86%Vals AIAs of 09/01/2026
Tax Agent BenchOverall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 8.8%Qwen 3.8 27B: 30.9%Vals AIAs of 10/26/2026
TaxEval v2Overall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 71.2%Qwen 3.8 27B: 70.9%Vals AIAs of 09/01/2026
Terminal-Bench 2.1Overall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 54.7%Qwen 3.8 27B: 58.4%Vals AIAs of 10/01/2026
Vibe Code Bench v1.1Overall · higher is betterTest setup not documented as identicalGPT-5.4 mini: 48%Qwen 3.8 27B: 64.9%Vals AIAs of 10/26/2026
Code MigrationCLI · higher is betterTest setup not documented as identicalGPT-5.4 mini: 8.2%Qwen 3.8 27B: 2.2%Vals AIAs of 10/18/2026
CritPt v2Overall · higher is betterGPT-5.4 mini: 0.0%Qwen 3.8 27B: 0.3%Artificial AnalysisAs of 10/29/2026
EMBComps · higher is betterTest setup not documented as identicalGPT-5.4 mini: 40.5%Qwen 3.8 27B: 53.3%Vals AIAs of 10/20/2026
Finance Agent (v2)Adjustments · higher is betterTest setup not documented as identicalGPT-5.4 mini: 45.4%Qwen 3.8 27B: 47.3%Vals AIAs of 10/20/2026
GDPval-AAOverall · higher is betterGPT-5.4 mini: 1,022.2 eloQwen 3.8 27B: 1,427.26 eloArtificial AnalysisAs of 10/29/2026
GPQA DiamondFew-Shot CoT · higher is betterTest setup not documented as identicalGPT-5.4 mini: 84.8%Qwen 3.8 27B: 88.9%Vals AIAs of 09/01/2026
Legal Research BenchAdministrative / Regulatory · All-pass · higher is betterTest setup not documented as identicalGPT-5.4 mini: 15.2%Qwen 3.8 27B: 45.5%Vals AIAs of 10/26/2026
LiveCodeBenchEasy · higher is betterTest setup not documented as identicalGPT-5.4 mini: 97.8%Qwen 3.8 27B: 99.1%Vals AIAs of 09/01/2026
MMLU ProBiology · higher is betterTest setup not documented as identicalGPT-5.4 mini: 91.4%Qwen 3.8 27B: 91.5%Vals AIAs of 09/01/2026
MMMU-ProOverall · higher is betterGPT-5.4 mini: 60.5%Qwen 3.8 27B: 69.9%Artificial AnalysisAs of 10/29/2026
MortgageTaxNumerical Extraction · higher is betterTest setup not documented as identicalGPT-5.4 mini: 79.4%Qwen 3.8 27B: 69.4%Vals AIAs of 09/01/2026
ProgramBenchAlmost Resolved · higher is betterTest setup not documented as identicalGPT-5.4 mini: 3%Qwen 3.8 27B: 1%Vals AIAs of 10/20/2026
SWE-bench Verified<15 min fix · higher is betterTest setup not documented as identicalGPT-5.4 mini: 85.6%Qwen 3.8 27B: 90.2%Vals AIAs of 09/01/2026
Tax Agent BenchControversy & Precedence Analysis · higher is betterTest setup not documented as identicalGPT-5.4 mini: 3.3%Qwen 3.8 27B: 6.7%Vals AIAs of 10/26/2026
TaxEval v2Correctness · higher is betterTest setup not documented as identicalGPT-5.4 mini: 63%Qwen 3.8 27B: 60.1%Vals AIAs of 09/01/2026
Terminal-Bench 2.1Easy · higher is betterTest setup not documented as identicalGPT-5.4 mini: 75%Qwen 3.8 27B: 50%Vals AIAs of 10/01/2026
Code MigrationCOBOL · higher is betterTest setup not documented as identicalGPT-5.4 mini: 27.1%Qwen 3.8 27B: 50%Vals AIAs of 10/18/2026
CritPt v2Overall · higher is betterGPT-5.4 mini: 10.0%Qwen 3.8 27B: 5.4%Artificial AnalysisAs of 10/29/2026
EMBDataroom · higher is betterTest setup not documented as identicalGPT-5.4 mini: 48.3%Qwen 3.8 27B: 79.7%Vals AIAs of 10/20/2026
Finance Agent (v2)All-Pass · higher is betterTest setup not documented as identicalGPT-5.4 mini: 32.4%Qwen 3.8 27B: 36.8%Vals AIAs of 10/20/2026
GPQA DiamondZero-Shot CoT · higher is betterTest setup not documented as identicalGPT-5.4 mini: 81.3%Qwen 3.8 27B: 88.9%Vals AIAs of 09/01/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit · higher is betterTest setup not documented as identicalGPT-5.4 mini: 69.7%Qwen 3.8 27B: 88.5%Vals AIAs of 10/26/2026
LiveCodeBenchHard · higher is betterTest setup not documented as identicalGPT-5.4 mini: 61.7%Qwen 3.8 27B: 65.7%Vals AIAs of 09/01/2026
MMLU ProBusiness · higher is betterTest setup not documented as identicalGPT-5.4 mini: 89.7%Qwen 3.8 27B: 90.2%Vals AIAs of 09/01/2026
MMMU-ProOverall · higher is betterGPT-5.4 mini: 73.3%Qwen 3.8 27B: 76.3%Artificial AnalysisAs of 10/29/2026
MortgageTaxSemantic Extraction · higher is betterTest setup not documented as identicalGPT-5.4 mini: 47.6%Qwen 3.8 27B: 60.5%Vals AIAs of 09/01/2026
ProgramBenchFully Resolved · higher is betterTest setup not documented as identicalGPT-5.4 mini: 0%Qwen 3.8 27B: 0%Vals AIAs of 10/20/2026
SWE-bench Verified>4 hours · higher is betterTest setup not documented as identicalGPT-5.4 mini: 0%Qwen 3.8 27B: 66.7%Vals AIAs of 09/01/2026
Tax Agent BenchControversy & Precedence Analysis · higher is betterTest setup not documented as identicalGPT-5.4 mini: 32.2%Qwen 3.8 27B: 48.3%Vals AIAs of 10/26/2026
TaxEval v2Stepwise Reasoning · higher is betterTest setup not documented as identicalGPT-5.4 mini: 79.4%Qwen 3.8 27B: 81.6%Vals AIAs of 09/01/2026
Terminal-Bench 2.1Hard · higher is betterTest setup not documented as identicalGPT-5.4 mini: 34.4%Qwen 3.8 27B: 43.3%Vals AIAs of 10/01/2026
Code MigrationCode Quality · higher is betterTest setup not documented as identicalGPT-5.4 mini: 3.053 / 5Qwen 3.8 27B: 4.197 / 5Vals AIAs of 10/18/2026
EMBDCF · higher is betterTest setup not documented as identicalGPT-5.4 mini: 40.6%Qwen 3.8 27B: 50%Vals AIAs of 10/20/2026
Finance Agent (v2)Comparables · higher is betterTest setup not documented as identicalGPT-5.4 mini: 40.3%Qwen 3.8 27B: 38.9%Vals AIAs of 10/20/2026
Legal Research BenchBusiness & Commercial · All-pass · higher is betterTest setup not documented as identicalGPT-5.4 mini: 11.6%Qwen 3.8 27B: 41.9%Vals AIAs of 10/26/2026
LiveCodeBenchMedium · higher is betterTest setup not documented as identicalGPT-5.4 mini: 84.9%Qwen 3.8 27B: 87.2%Vals AIAs of 09/01/2026
MMLU ProChemistry · higher is betterTest setup not documented as identicalGPT-5.4 mini: 89.1%Qwen 3.8 27B: 91.8%Vals AIAs of 09/01/2026
ProgramBenchRaw Pass Rate · higher is betterTest setup not documented as identicalGPT-5.4 mini: 54.4%Qwen 3.8 27B: 11.2%Vals AIAs of 10/20/2026
SWE-bench Verified1-4 hours · higher is betterTest setup not documented as identicalGPT-5.4 mini: 33.3%Qwen 3.8 27B: 69%Vals AIAs of 09/01/2026
Tax Agent BenchCurrent & Temporal Analysis · higher is betterTest setup not documented as identicalGPT-5.4 mini: 10%Qwen 3.8 27B: 43.1%Vals AIAs of 10/26/2026
Terminal-Bench 2.1Medium · higher is betterTest setup not documented as identicalGPT-5.4 mini: 64.2%Qwen 3.8 27B: 67.3%Vals AIAs of 10/01/2026
EMBLBO · higher is betterTest setup not documented as identicalGPT-5.4 mini: 41.6%Qwen 3.8 27B: 53.7%Vals AIAs of 10/20/2026
Finance Agent (v2)Disclosure Analysis · higher is betterTest setup not documented as identicalGPT-5.4 mini: 48.3%Qwen 3.8 27B: 60.3%Vals AIAs of 10/20/2026
Legal Research BenchBusiness & Commercial · Weighted partial credit · higher is betterTest setup not documented as identicalGPT-5.4 mini: 71.2%Qwen 3.8 27B: 86.8%Vals AIAs of 10/26/2026
MMLU ProComputer Science · higher is betterTest setup not documented as identicalGPT-5.4 mini: 87.1%Qwen 3.8 27B: 88.3%Vals AIAs of 09/01/2026
SWE-bench Verified15 min - 1 hour · higher is betterTest setup not documented as identicalGPT-5.4 mini: 70.9%Qwen 3.8 27B: 85.8%Vals AIAs of 09/01/2026
Compare another matchup
More matchups
Related LLM comparisons
Compare GPT-5.4 mini and Qwen 3.8 27B with other leading models.