Kimi K3vs.Qwen 3.8 27B
Which one is better?
Kimi K3 is clearly ahead in the overall index, but costs 75 times (input) and 6 times (output) as much per token.
59
Kimi K3
Rank 18 to 24
11
Qwen 3.8 27B
Rank 44 to 57
55
Kimi K3
slow
40
Qwen 3.8 27B
slow
$6
Kimi K3
expensive
$0.66
Qwen 3.8 27B
fairly cheap
Kimi K3
- Subscription
At Moonshot AI from ≈ €20.14 a month (Kimi Plus, with limited credits). Through other providers from ≈ €1.06 a month (Command Code Go, with limited credits). 45 plans in Germany.
- Via API
$3 input, $15 output per 1M tokens, unchanged since our price tracking began on Aug 14, 2026.
- Note
- There is a successor: Kimi K3.1 Pro (rank 10 to 13, $2 input, $10 output).
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.
Kimi K3 ahead: 9Qwen 3.8 27B ahead: 0level: 0too little data: 1of 10 areas
AreaKimi K3Qwen 3.8 27B0 to 100, higher is betterResult
Areas in the overall index
CodingKimi K3 aheadKimi K3: 48Qwen 3.8 27B: 13
ReasoningKimi K3 aheadKimi K3: 53Qwen 3.8 27B: 6
MathematicsKimi K3 aheadKimi K3: 55Qwen 3.8 27B: 8
KnowledgeKimi K3 aheadKimi K3: 44Qwen 3.8 27B: 5
AgentsKimi K3 aheadKimi K3: 49Qwen 3.8 27B: 21
Generalonly Kimi K3 has a scoreKimi K3: 72Qwen 3.8 27B: not enough data
MultimodalKimi K3 aheadKimi K3: 54Qwen 3.8 27B: 13
Specialist fields
FinanceKimi K3 aheadKimi K3: 63Qwen 3.8 27B: 34
LegalKimi K3 aheadKimi K3: 74Qwen 3.8 27B: 23
HealthKimi K3 aheadKimi K3: 65Qwen 3.8 27B: 3
Show 3 more areas that do not count toward the overall index
AreaKimi K3Qwen 3.8 27B0 to 100, higher is betterResult
CybersecurityKimi K3 aheadKimi K3: 75Qwen 3.8 27B: 19
Productivityonly Kimi K3 has a scoreKimi K3: 71Qwen 3.8 27B: not enough data
Educationonly Kimi K3 has a scoreKimi K3: 55Qwen 3.8 27B: not enough data
See for yourself: same task, one attempt
SVG short test
One SVG task with the same fixed settings for every model, without hidden thinking.
Shared test round: 43b8c5c46830
Kimi K3
Test details
API reports 0 reasoning tokensAPI model: moonshotai/kimi-k3495 output tokensQwen 3.8 27B
Test details
API reports 0 reasoning tokensAPI model: qwen/qwen3.8-27b865 output tokensEarlier tests
Pelican on a bicycle
A community classic for free-form SVG drawing.
Kimi K3
Reasoning (requested): high
Current attempt
- Run date
- Oct 1, 2026
- API provider
- openrouter
- Requested API model
- moonshotai/kimi-k3
- API model reported in response
- moonshotai/kimi-k3
- Reasoning setting
- high
- Pinned endpoint
- Moonshot AI | moonshotai/kimi-k3-20260715
- 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
- 173.4 s
Visually reviewed. This individual test does not establish overall model quality.
Earlier attempts (1)
- 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)All data side by side
The better value is bold. For the overall index only when the gap is clear.
| Metric | Kimi K3 | Qwen 3.8 27B |
|---|---|---|
| Intelligence | ||
| Overall index0 to 100, higher is better | 59 | 11 |
| Rank | Rank 18 to 24 of 95 | Rank 44 to 57 of 95 |
| Price per 1M tokens | ||
| Inputwhat you send | $3 | $0.04 |
| Outputwhat the model writes | $15 | $2.50 |
| Speed | ||
| Tokens per second | 55 | 40 |
| Technical | ||
| Context windowhow much text the model reads at once | 1.05 M tokens≈ 1,570 A4 pages | 262,144 tokens≈ 390 A4 pages |
| Knowledge cutoff | n/a | n/a |
| Released | Jul 16, 2026 | Aug 14, 2026 |
| Availability | Active | Active |
| Open weights | yes | 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: Kimi K3B: Qwen 3.8 27B
AA-Omniscience AccuracyOverall · higher is betterKimi K3: 42.4%Qwen 3.8 27B: 18.3%Artificial AnalysisAs of 10/29/2026
AA-Omniscience Hallucination RateOverall · lower is betterKimi K3: 76.6%Qwen 3.8 27B: 48.7%Artificial AnalysisAs of 10/29/2026
Code MigrationOverall · higher is betterTest setup not documented as identicalKimi K3: 16.1%Qwen 3.8 27B: 14.2%Vals AIAs of 10/18/2026
CritPt v2Overall · higher is betterKimi K3: 3.1%Qwen 3.8 27B: 0.0%Artificial AnalysisAs of 10/29/2026
CyberBench v1.1Overall · higher is betterTest setup not documented as identicalKimi K3: 75.2%Qwen 3.8 27B: 67.8%Vals AIAs of 10/20/2026
EMBOverall · higher is betterTest setup not documented as identicalKimi K3: 66.7%Qwen 3.8 27B: 59.7%Vals AIAs of 10/20/2026
Finance Agent (v2)Overall · higher is betterTest setup not documented as identicalKimi K3: 53.1%Qwen 3.8 27B: 48.6%Vals AIAs of 10/20/2026
FrontierCode 1.1 MainOverall · higher is betterKimi K3: 28.5%Qwen 3.8 27B: 28.9%Artificial AnalysisAs of 10/29/2026
GDPval-AAOverall · higher is betterKimi K3: 1,136.21 eloQwen 3.8 27B: 1,370.88 eloArtificial AnalysisAs of 10/29/2026
GPQA DiamondOverall · higher is betterTest setup not documented as identicalKimi K3: 92.9%Qwen 3.8 27B: 88.9%Vals AIAs of 09/01/2026
Show 54 more benchmarks
IOIOverall · higher is betterTest setup not documented as identicalKimi K3: 48.9%Qwen 3.8 27B: 39.1%Vals AIAs of 10/24/2026
Legal Research BenchOverall · All-pass · higher is betterTest setup not documented as identicalKimi K3: 46.2%Qwen 3.8 27B: 36.1%Vals AIAs of 10/26/2026
LegalBenchOverall · higher is betterTest setup not documented as identicalKimi K3: 86.2%Qwen 3.8 27B: 82.4%Vals AIAs of 10/13/2026
LiveCodeBenchOverall · higher is betterTest setup not documented as identicalKimi K3: 87.2%Qwen 3.8 27B: 84%Vals AIAs of 09/01/2026
MedCodeOverall · higher is betterTest setup not documented as identicalKimi K3: 49.4%Qwen 3.8 27B: 28.7%Vals AIAs of 10/13/2026
MedScribeOverall · higher is betterTest setup not documented as identicalKimi K3: 88%Qwen 3.8 27B: 83.8%Vals AIAs of 10/13/2026
MMLU ProOverall · higher is betterTest setup not documented as identicalKimi K3: 88%Qwen 3.8 27B: 84.3%Vals AIAs of 09/01/2026
MMMU ProOverall · higher is betterTest setup not documented as identicalKimi K3: 88.2%Qwen 3.8 27B: 83.9%Vals AIAs of 09/01/2026
MMMU-ProOverall · higher is betterKimi K3: 78.5%Qwen 3.8 27B: 73.8%Artificial AnalysisAs of 10/29/2026
MortgageTaxOverall · higher is betterTest setup not documented as identicalKimi K3: 66.3%Qwen 3.8 27B: 64.9%Vals AIAs of 09/01/2026
ProgramBenchOverall · higher is betterTest setup not documented as identicalKimi K3: 2%Qwen 3.8 27B: 0%Vals AIAs of 10/20/2026
ProofBench v1.1Overall · higher is betterTest setup not documented as identicalKimi K3: 87%Qwen 3.8 27B: 16%Vals AIAs of 10/24/2026
SWE-bench VerifiedOverall · higher is betterTest setup not documented as identicalKimi K3: 93.4%Qwen 3.8 27B: 86%Vals AIAs of 09/01/2026
Tax Agent BenchOverall · higher is betterTest setup not documented as identicalKimi K3: 33.4%Qwen 3.8 27B: 30.9%Vals AIAs of 10/26/2026
TaxEval v2Overall · higher is betterTest setup not documented as identicalKimi K3: 75.7%Qwen 3.8 27B: 70.9%Vals AIAs of 09/01/2026
Terminal-Bench 2.1Overall · higher is betterTest setup not documented as identicalKimi K3: 80.9%Qwen 3.8 27B: 58.4%Vals AIAs of 10/01/2026
Terminal-Bench-ScienceOverall · higher is betterTest setup not documented as identicalKimi K3: 1.4%Qwen 3.8 27B: 0%Vals AIAs of 10/25/2026
Vibe Code Bench v1.1Overall · higher is betterTest setup not documented as identicalKimi K3: 85%Qwen 3.8 27B: 64.9%Vals AIAs of 10/26/2026
Code MigrationCLI · higher is betterTest setup not documented as identicalKimi K3: 1.5%Qwen 3.8 27B: 2.2%Vals AIAs of 10/18/2026
CyberBench v1.1Patch · higher is betterTest setup not documented as identicalKimi K3: 82.1%Qwen 3.8 27B: 83.9%Vals AIAs of 10/20/2026
EMBComps · higher is betterTest setup not documented as identicalKimi K3: 70.8%Qwen 3.8 27B: 53.3%Vals AIAs of 10/20/2026
Finance Agent (v2)Adjustments · higher is betterTest setup not documented as identicalKimi K3: 45.3%Qwen 3.8 27B: 47.3%Vals AIAs of 10/20/2026
GPQA DiamondFew-Shot CoT · higher is betterTest setup not documented as identicalKimi K3: 93.4%Qwen 3.8 27B: 88.9%Vals AIAs of 09/01/2026
IOIIOI 2024 · higher is betterTest setup not documented as identicalKimi K3: 59.7%Qwen 3.8 27B: 31.8%Vals AIAs of 10/24/2026
Legal Research BenchAdministrative / Regulatory · All-pass · higher is betterTest setup not documented as identicalKimi K3: 63.6%Qwen 3.8 27B: 45.5%Vals AIAs of 10/26/2026
LegalBenchConclusion Tasks · higher is betterTest setup not documented as identicalKimi K3: 90.1%Qwen 3.8 27B: 88.4%Vals AIAs of 10/13/2026
LiveCodeBenchEasy · higher is betterTest setup not documented as identicalKimi K3: 98.8%Qwen 3.8 27B: 99.1%Vals AIAs of 09/01/2026
MMLU ProBiology · higher is betterTest setup not documented as identicalKimi K3: 93.9%Qwen 3.8 27B: 91.5%Vals AIAs of 09/01/2026
MortgageTaxNumerical Extraction · higher is betterTest setup not documented as identicalKimi K3: 78.9%Qwen 3.8 27B: 69.4%Vals AIAs of 09/01/2026
ProgramBenchAlmost Resolved · higher is betterTest setup not documented as identicalKimi K3: 17.5%Qwen 3.8 27B: 1%Vals AIAs of 10/20/2026
SWE-bench Verified<15 min fix · higher is betterTest setup not documented as identicalKimi K3: 92.3%Qwen 3.8 27B: 90.2%Vals AIAs of 09/01/2026
Tax Agent BenchControversy & Precedence Analysis · higher is betterTest setup not documented as identicalKimi K3: 13.3%Qwen 3.8 27B: 6.7%Vals AIAs of 10/26/2026
TaxEval v2Correctness · higher is betterTest setup not documented as identicalKimi K3: 65.3%Qwen 3.8 27B: 60.1%Vals AIAs of 09/01/2026
Terminal-Bench 2.1Easy · higher is betterTest setup not documented as identicalKimi K3: 91.7%Qwen 3.8 27B: 50%Vals AIAs of 10/01/2026
Terminal-Bench-ScienceEarth Sciences · higher is betterTest setup not documented as identicalKimi K3: 12.5%Qwen 3.8 27B: 0%Vals AIAs of 10/25/2026
Code MigrationCOBOL · higher is betterTest setup not documented as identicalKimi K3: 60%Qwen 3.8 27B: 50%Vals AIAs of 10/18/2026
CyberBench v1.1PoC · higher is betterTest setup not documented as identicalKimi K3: 68.3%Qwen 3.8 27B: 51.7%Vals AIAs of 10/20/2026
EMBDataroom · higher is betterTest setup not documented as identicalKimi K3: 82.7%Qwen 3.8 27B: 79.7%Vals AIAs of 10/20/2026
Finance Agent (v2)All-Pass · higher is betterTest setup not documented as identicalKimi K3: 41.8%Qwen 3.8 27B: 36.8%Vals AIAs of 10/20/2026
GPQA DiamondZero-Shot CoT · higher is betterTest setup not documented as identicalKimi K3: 92.4%Qwen 3.8 27B: 88.9%Vals AIAs of 09/01/2026
IOIIOI 2025 · higher is betterTest setup not documented as identicalKimi K3: 29.3%Qwen 3.8 27B: 48.5%Vals AIAs of 10/24/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit · higher is betterTest setup not documented as identicalKimi K3: 89.3%Qwen 3.8 27B: 88.5%Vals AIAs of 10/26/2026
LegalBenchInterpretation Tasks · higher is betterTest setup not documented as identicalKimi K3: 84.1%Qwen 3.8 27B: 82%Vals AIAs of 10/13/2026
LiveCodeBenchHard · higher is betterTest setup not documented as identicalKimi K3: 74.3%Qwen 3.8 27B: 65.7%Vals AIAs of 09/01/2026
MMLU ProBusiness · higher is betterTest setup not documented as identicalKimi K3: 92%Qwen 3.8 27B: 90.2%Vals AIAs of 09/01/2026
MortgageTaxSemantic Extraction · higher is betterTest setup not documented as identicalKimi K3: 53.8%Qwen 3.8 27B: 60.5%Vals AIAs of 09/01/2026
ProgramBenchFully Resolved · higher is betterTest setup not documented as identicalKimi K3: 2%Qwen 3.8 27B: 0%Vals AIAs of 10/20/2026
SWE-bench Verified>4 hours · higher is betterTest setup not documented as identicalKimi K3: 66.7%Qwen 3.8 27B: 66.7%Vals AIAs of 09/01/2026
Tax Agent BenchControversy & Precedence Analysis · higher is betterTest setup not documented as identicalKimi K3: 68.8%Qwen 3.8 27B: 48.3%Vals AIAs of 10/26/2026
TaxEval v2Stepwise Reasoning · higher is betterTest setup not documented as identicalKimi K3: 86.1%Qwen 3.8 27B: 81.6%Vals AIAs of 09/01/2026
Terminal-Bench 2.1Hard · higher is betterTest setup not documented as identicalKimi K3: 74.4%Qwen 3.8 27B: 43.3%Vals AIAs of 10/01/2026
Terminal-Bench-ScienceEngineering Sciences · higher is betterTest setup not documented as identicalKimi K3: 0%Qwen 3.8 27B: 0%Vals AIAs of 10/25/2026
Code MigrationCode Quality · higher is betterTest setup not documented as identicalKimi K3: 3.899 / 5Qwen 3.8 27B: 4.197 / 5Vals AIAs of 10/18/2026
EMBDCF · higher is betterTest setup not documented as identicalKimi K3: 56.7%Qwen 3.8 27B: 50%Vals AIAs of 10/20/2026