{"name":"Qwen-3.8 27B","arena":{"id":"qwen-3-8-27b-bf16","name":"Qwen-3.8 27B","vram":54,"params":27,"vendor":"alibaba","quality":{"coding":{"bench":"LiveCodeBench v6","value":90.3},"science":{"bench":"GPQA-Diamond","value":89.2},"knowledge":null},"contextK":256,"baseModelId":"qwen-3-8-27b"},"notes":"Model-ID: Qwen/Qwen3.8-27B. Container: vllm/vllm-openai:v0.26.0-aarch64-cu129-ubuntu2404. llama-benchy 0.4.0. Bronrun: 2026-08-15T10:57:03+02:00. KV-cache fp8; max_model_len 131072; gpu_memory_utilization 0,9. Prefix caching uit; async scheduling aan. Extra serverflags: --language-model-only --reasoning-parser qwen3 --trust-remote-code. 11 van de 11 tests hebben een meting die volgens de huidige Arena-regels meetelt. De sanity-check is geslaagd. Dit is een controle op een kort gegenereerd antwoord, geen inhoudelijke evaluatie van de benchmarktaken.","order":370,"engine":"vLLM v0.26.0","verdict":"Gemeten op 2026-08-15 met vLLM v0.26.0. In de chattest met tien gelijktijdige verzoeken haalt dit profiel 3,99 tokens/s per gebruiker, met gemiddeld 6,09 seconden tot het eerste token. Bij 25k context stijgt die gemiddelde wachttijd naar 119,42 seconden. Deze runs meten snelheid en wachttijd, niet de kwaliteit van de antwoorden.","hardware":"DGX Spark, NVIDIA GB10, 128 GB unified memory","modelUrl":"https://huggingface.co/Qwen/Qwen3.8-27B","provider":"Qwen","precision":"BF16","parameters":"27B","provenance":{"modelId":"Qwen/Qwen3.8-27B","runPath":"results/qwen-3.8/qwen-3.8-27b/bf16","validity":{"tests":{"01-chat":"complete","02-rag-8k":"complete","08-sharegpt":"complete","09-reasoning":"complete","04-multi-turn":"complete","11-rate-sweep":"complete","03-long-output":"complete","05-big-context":"complete","10-monday-peak":"complete","07-office-baseline":"complete","06-concurrency-stress":"complete"},"sanity":"passed"},"sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","servedName":"qwen3.8-27b-bf16","generatedAt":"2026-08-15T10:57:03+02:00","runRevision":"a948cec23b054c62fef595bad3cdb284066922ac","vllmVersion":"0.26.0","suiteVersion":"2026-08","repositoryUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks","contractVersion":1,"contractRevision":"a948cec23b054c62fef595bad3cdb284066922ac","commandProvenance":"reconstructed_from_raw_artifacts","llamaBenchyVersion":"0.4.0"},"architecture":"Dense","measuredContextK":128,"benchmarkResults":[{"id":"qwen-3-8-27b-bf16-chat-1k","date":"2026-08-15","runs":3,"metrics":{"ttfrSec":{"mean":6.086,"stddev":2.265},"decodeTokensPerSecondTotal":{"mean":39.14,"stddev":0.04},"prefillTokensPerSecondTotal":{"mean":1069.15,"stddev":24.7},"decodeTokensPerSecondPerUser":{"mean":3.99,"stddev":0.03}},"sourceRun":"Test 01","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","workloadId":"chat-1k","sourceLabel":"Qwen/Qwen3.8-27B bf16 op de DGX Spark","sourceStatus":"complete","sourceTestId":"01-chat","prefixCaching":false},{"id":"qwen-3-8-27b-bf16-rag-8k","date":"2026-08-15","runs":3,"metrics":{"ttfrSec":{"mean":37.787,"stddev":18.872},"decodeTokensPerSecondTotal":{"mean":27.16,"stddev":0.03},"prefillTokensPerSecondTotal":{"mean":1118.96,"stddev":1.32},"decodeTokensPerSecondPerUser":{"mean":3.35,"stddev":0.36}},"sourceRun":"Test 02","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","workloadId":"rag-8k","sourceLabel":"Qwen/Qwen3.8-27B bf16 op de DGX Spark","sourceStatus":"complete","sourceTestId":"02-rag-8k","prefixCaching":false},{"id":"qwen-3-8-27b-bf16-long-output-agents","date":"2026-08-15","runs":3,"metrics":{"ttfrSec":{"mean":2.141,"stddev":0.359},"decodeTokensPerSecondTotal":{"mean":37.66,"stddev":0.93},"prefillTokensPerSecondTotal":{"mean":907.92,"stddev":7.63},"decodeTokensPerSecondPerUser":{"mean":4.07,"stddev":0.01}},"sourceRun":"Test 03","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","workloadId":"long-output-agents","sourceLabel":"Qwen/Qwen3.8-27B bf16 op de DGX Spark","sourceStatus":"complete","sourceTestId":"03-long-output","prefixCaching":false},{"id":"qwen-3-8-27b-bf16-multi-turn-office","date":"2026-08-15","runs":3,"metrics":{"ttfrSec":{"mean":10.259,"stddev":4.821},"decodeTokensPerSecondTotal":{"mean":36.39,"stddev":0.03},"prefillTokensPerSecondTotal":{"mean":1135.66,"stddev":4.93},"decodeTokensPerSecondPerUser":{"mean":3.89,"stddev":0.12}},"sourceRun":"Test 04","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","workloadId":"multi-turn-office","sourceLabel":"Qwen/Qwen3.8-27B bf16 op de DGX Spark","sourceStatus":"complete","sourceTestId":"04-multi-turn","prefixCaching":false},{"id":"qwen-3-8-27b-bf16-large-context-25k","date":"2026-08-15","runs":3,"metrics":{"ttfrSec":{"mean":119.421,"stddev":61.568},"decodeTokensPerSecondTotal":{"mean":9.82,"stddev":0.03},"prefillTokensPerSecondTotal":{"mean":1053.52,"stddev":0.48},"decodeTokensPerSecondPerUser":{"mean":1.96,"stddev":0.83}},"sourceRun":"Test 05","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","workloadId":"large-context-25k","sourceLabel":"Qwen/Qwen3.8-27B bf16 op de DGX Spark","sourceStatus":"complete","sourceTestId":"05-big-context","prefixCaching":false},{"id":"qwen-3-8-27b-bf16-concurrency-stress-25k","date":"2026-08-15","runs":3,"metrics":{"ttfrSec":{"mean":228.576,"stddev":125.763},"decodeTokensPerSecondTotal":{"mean":10.4,"stddev":0.03},"prefillTokensPerSecondTotal":{"mean":1037.12,"stddev":0.12},"decodeTokensPerSecondPerUser":{"mean":1.26,"stddev":0.73}},"sourceRun":"Test 06","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","workloadId":"concurrency-stress-25k","sourceLabel":"Qwen/Qwen3.8-27B bf16 op de DGX Spark","sourceStatus":"complete","sourceTestId":"06-concurrency-stress","prefixCaching":false}],"openLoopResults":[{"id":"qwen-3-8-27b-bf16-office-random-4k","date":"2026-08-15","metrics":{"e2eMs":{"p50":627365.84,"p90":924354.75,"p95":993265.69,"p99":1057341.85,"mean":626520.39},"tpotMs":{"p50":1271.1,"p90":2004.16,"p95":2046.74,"p99":2102.37,"mean":1290.54},"ttftMs":{"p50":87195.24,"p90":151876.13,"p95":159985.8,"p99":181100.66,"mean":86257.9}},"sourceRun":"Test 07","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","scenarioId":"office-random-4k","achievedRps":0.162,"sourceStatus":"complete","sourceTestId":"07-office-baseline","configuredRps":0.3,"prefixCaching":false,"totalRequests":200,"successfulRequests":200,"totalTokenThroughput":747.26,"peakConcurrentRequests":175},{"id":"qwen-3-8-27b-bf16-sharegpt-replay","date":"2026-08-15","metrics":{"e2eMs":{"p50":43541.95,"p90":169503.61,"p95":214353.05,"p99":282161.35,"mean":69438.87},"tpotMs":{"p50":341.48,"p90":358.12,"p95":372.22,"p99":389.21,"mean":330.1},"ttftMs":{"p50":1008.21,"p90":1494.92,"p95":1599.18,"p99":1916.34,"mean":1072.41}},"sourceRun":"Test 08","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","scenarioId":"sharegpt-replay","achievedRps":0.261,"sourceStatus":"complete","sourceTestId":"08-sharegpt","configuredRps":0.3,"prefixCaching":false,"totalRequests":250,"successfulRequests":250,"totalTokenThroughput":124.82,"peakConcurrentRequests":34},{"id":"qwen-3-8-27b-bf16-reasoning","date":"2026-08-15","metrics":{"e2eMs":{"p50":1455660.58,"p90":2263712.96,"p95":2433987.04,"p99":2487832.65,"mean":1504049.94},"tpotMs":{"p50":370.94,"p90":379.81,"p95":382.86,"p99":387.58,"mean":365.28},"ttftMs":{"p50":1797.45,"p90":2624.52,"p95":2874.47,"p99":3157.65,"mean":1889.96}},"sourceRun":"Test 09","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","scenarioId":"reasoning","achievedRps":0.019,"sourceStatus":"complete","sourceTestId":"09-reasoning","configuredRps":0.2,"prefixCaching":false,"totalRequests":50,"successfulRequests":50,"totalTokenThroughput":102.01,"peakConcurrentRequests":50},{"id":"qwen-3-8-27b-bf16-monday-peak-random-4k","date":"2026-08-15","metrics":{"e2eMs":{"p50":237886.93,"p90":430941.2,"p95":448648.23,"p99":483722.27,"mean":245847.86},"tpotMs":{"p50":481.78,"p90":520.4,"p95":549.15,"p99":638.34,"mean":481.3},"ttftMs":{"p50":5393.8,"p90":10589.83,"p95":28039.69,"p99":73914.67,"mean":8348.08}},"sourceRun":"Test 10","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","scenarioId":"monday-peak-random-4k","achievedRps":0.098,"sourceStatus":"complete","sourceTestId":"10-monday-peak","configuredRps":1.5,"prefixCaching":false,"totalRequests":300,"successfulRequests":300,"totalTokenThroughput":446.79,"peakConcurrentRequests":29}],"rateSweep":{"date":"2026-08-15","capacity":{"slo2000":null,"slo5000":null,"slo10000":null},"sourceRun":"Test 11","sourceUrl":"https://github.com/djangodevreng/dgx-spark-benchmarks/tree/a948cec23b054c62fef595bad3cdb284066922ac/results/qwen-3.8/qwen-3.8-27b/bf16/","abortedAtRps":null,"queueKneeRps":null,"sourceStatus":"complete","sourceTestId":"11-rate-sweep","totalRequests":850,"successfulRequests":850}}