Which is better, Qwen3.5 122B A10B or Qwen3.5 397B A17B?

Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, Qwen3.5 122B A10B or Qwen3.5 397B A17B?: Across 54 shared benchmarks, Qwen3.5 122B A10B scores higher on 7 and Qwen3.5 397B A17B on 46, with 1 level. The widest gap is AA Agentic Index, where Qwen3.5 397B A17B scores 53.3 against 21.3. Qwen3.5 122B A10B is the cheaper of the two on tracked API pricing ($0.40 against $0.60 per million input tokens).

Qwen3.5 122B A10B vs Qwen3.5 397B A17B

Across 54 shared benchmarks, Qwen3.5 122B A10B scores higher on 7 and Qwen3.5 397B A17B on 46, with 1 level. The widest gap is AA Agentic Index, where Qwen3.5 397B A17B scores 53.3 against 21.3. Qwen3.5 122B A10B is the cheaper of the two on tracked API pricing ($0.40 against $0.60 per million input tokens).

AlibabavsAlibaba54 shared benchmarks7–46 head-to-head

Benchmark	Qwen3.5 122B A10B	Qwen3.5 397B A17B
AA Agentic Index	21.3	53.3
AA Intelligence	32.8	34.3
AA-LCR	70.3	72.7
AA-Omniscience	-41.5	-30.8
arena_vision	1246	1265
Artificial Analysis Coding Index	45.7	48.2
bfcl_v4	72.2	72.9
browsecomp	63.8	69
browsecomp_zh	69.9	70.3
C-Eval	91.9	93
CC-OCR	81.8	82
coding_arena_elo	1358	1399
critpt	0.9	1.7
DeepPlanning	24.1	34.3
DynaMath	85.9	86.3
ERQA	62	67.5
gdpval	24.3	35.8
GPQA Diamond	86.6	89.3
GSM8K	94.5	66.7
HLE	47.5	29
HMMT 2025	90.3	92.7
IFBench	76.1	78.8
ifeval	93.4	92.6
include	82.8	85.6
LiveCodeBench v6	78.9	83.6
longbench_v2	60.2	63.2
MLVU	87.3	86.7
mmlu_prox	82.2	84.7
mmlu_redux	94	94.9
MMLU-Pro	86.7	87.8
mmmlu	86.7	88.5
MMMU	83.9	85
MMMU-Pro	76.9	79
MMStar	82.9	83.8
multichallenge	61.5	67.6
MV-Bench	76.6	77.6
OmniScience Accuracy	24.4	30.8
OmniScience Non-Hallucination	12.9	17.3
RealWorldQA	85.1	83.9
scicode	42	42
Seal-0	44.1	46.9
SimpleVQA	0.6	67.1
supergpqa	67.1	70.4
SWE-bench Verified	72	76.4
t2-bench	79.5	86.7
TauBench V3 - Banking	15.3	13.4
Terminal-Bench 2.0	49.4	52.5
Terminal-Bench 2.1	47.6	51.3
Terminal-Bench Hard	31.1	40.9
VideoMMMU	82	84.7
VITA-Bench	33.6	49.7
WideSearch	60.5	74
τ²-Bench Telecom (AA run)	93.6	95.6
τ³-Bench	13.6	13.4

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, alibaba-official), otherwise the lowest tracked offer.