Qwen3.6 35B A3B
- Input
- $0.375/M tok
- Output
- $2.25/M tok
- Blended
- $0.844/M tok · 3:1
- Speed
- 120tok/s
- Context
- 262Ktokens
- Cost per task
- ~$0.95/taskest.
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Qwen3.6 35B A3B (Reasoning). Price = AA median host price.
Rankings
Coding
Verified#106 / 137rank by quality30.8±8.1quality / 1002/7 benchmarks measuredWriting & chat
Verified#72 / 162rank by quality50.1±9.5quality / 1001/5 benchmarks measuredVision
Verified#61 / 98rank by quality47.6±6.3quality / 1002/6 benchmarks measuredHard reasoning
Verified#97 / 218rank by quality48.7±3.7quality / 1003/4 benchmarks measuredAgents
Verified#89 / 187rank by quality34.1±4.3quality / 1004/7 benchmarks measured
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs, compared with the API price above.
- Hardware
- 1× H200 141GB
- FP8 weights
- Throughput
- ~8,390 tok/s
- many requests batched
- Price
- $0.046–0.38 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, 300+ tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (35B params, 3B active per token) + 25% KV-cache headroom ≈ 44 GB
- 1× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
- ~8,390 output tok/s aggregate at batch 256 (bandwidth-bound); MoE compute scales with active params
- Blended 3:1 input:output; prefill ~155,846 tok/s
- Low = 75% utilization at the low GPU price; high = 20% at the high price
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedEstimated
~#17 of 21 (estimated)~69.3%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #16Nemotron 3 Ultra70.7%
- ~#17Qwen3.6 35B A3BEstimated69.3%
- #17Intern-S2-Preview (35B-A3B)64.0%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#23 of 24 (estimated)~48.4%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #22Muse Glimmer51.2%
- ~#23Qwen3.6 35B A3BEstimated48.4%
- #23Nex-N2.5-Mini43.8%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#40 of 46 (estimated)~42.2%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #39Beam 1.0 501B A23B43.3%
- ~#40Qwen3.6 35B A3BEstimated42.2%
- #40Qwen3.8 27B42.2%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#17 of 28 (estimated)~60.9%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #16Claude Haiku 4.561.5%
- ~#17Qwen3.6 35B A3BEstimated60.9%
- #17GLM-5.3 NVFP453.5%
No published score — estimated from related benchmarks
SciCodeVerified
#92 of 11836.6%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #92Mistral Large 336.6%
- #92Qwen3.6 35B A3B36.6%
- #94LongCat 2.036.3%
artificialanalysis.ai · 2026-09-24
AA Coding Agent IndexEstimated
~#25 of 26 (estimated)~29.6
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #24GPT-6 Luna31.1
- ~#25Qwen3.6 35B A3BEstimated29.6
- #25Muse Spark Open26.3
No published score — estimated from related benchmarks
Writing & chat
IFBenchVerified
#52 of 13964.4%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #51gpt-oss-20b65.1%
- #52Qwen3.6 35B A3B64.4%
- #53Nemotron 3 Nano Omni 30B A3B Reasoning63.2%
artificialanalysis.ai · 2026-09-24
LMArena Text, Non-English (Elo)Estimated
~#42 of 57 (estimated)~1391 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #41Claude Haiku 4.51395
- ~#42Qwen3.6 35B A3BEstimated1391
- #42GPT-5.4 nano1387
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#42 of 57 (estimated)~1408 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #41Mistral Large 31413
- ~#42Qwen3.6 35B A3BEstimated1408
- #42GPT-5.4 nano1402
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#19 of 26 (estimated)~1609 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #18Inkling1611
- ~#19Qwen3.6 35B A3BEstimated1609
- #19Gemini 3.5 Flash-Lite1559
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#37 of 8775.0%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #37Qwen3.5 122B A10B75.0%
- #37Qwen3.6 35B A3B75.0%
- #39Qwen3.8 27B NVFP474.9%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~#11 of 12 (estimated)~76.1%
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #10Claude Sonnet 577.0%
- ~#11Qwen3.6 35B A3BEstimated76.1%
- #11Gemini 3.5 Flash-Lite74.5%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#5 of 8 (estimated)~82.1%
- #1MiniMax-M391.6%
- #2Kimi K391.1%
- #4Gemini 3.1 Pro85.3%
- ~#5Qwen3.6 35B A3BEstimated82.1%
- #5Claude Haiku 4.579.6%
No published score — estimated from related benchmarks
GDP.pdf (AA)Verified
#37 of 476.6%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #36GPT-5.4 nano7.8%
- #37Qwen3.6 35B A3B6.6%
- #38Claude Haiku 4.53.8%
artificialanalysis.ai · 2026-09-24
ChartographyEstimated
~#22 of 25 (estimated)~10.8%
- #1Claude Fable 5.273.4%
- #2GPT-6 Astra71.0%
- #21DeepSeek V4.1 Flash11.6%
- ~#22Qwen3.6 35B A3BEstimated10.8%
- #22Gemini 3.5 Flash-Lite10.2%
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#63 of 19284.1%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #61Nanbeige4.1-3B84.9%
- #63Qwen3.6 35B A3B84.1%
- #64Gemini 3.5 Flash-Lite83.8%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#73 of 19822.2%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #72Ling-3.0-flash-Fin22.6%
- #73Qwen3.6 35B A3B22.2%
- #74Ling-3.0-flash-VL22.0%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#19 of 34 (estimated)~85.9%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #18Mistral Medium 3.586.3%
- ~#19Qwen3.6 35B A3BEstimated85.9%
- #19Mistral Small 484.0%
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#5 of 11995.3%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #3Qwen3.5 397B A17B95.6%
- #5Qwen3.6 35B A3B95.3%
- #6Mistral Medium 3.594.2%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~63.4%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Qwen3.6 35B A3BEstimated63.4%
- #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#19 of 22 (estimated)~61.2%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19Qwen3.6 35B A3BEstimated61.2%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#47 of 91879 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #46GPT-5.4 nano937
- #47Qwen3.6 35B A3B879
- #48K2 Horizon 3.7B865
artificialanalysis.ai · 2026-09-24