MDL
Qwen3.5 122B A10B
Alibaba
Verified against Artificial Analysis · Jul 13, 2026
Capability
87
percentile index
Price /M
$0.4 / $3.2
in / out
Context
262.1K
tokens
Avg score
59.7
7 benchmarks
AA Intelligence
32.3
index
Speed
147 tok/s
TTFT 1.1s
Released
Feb 24, 2026
About
Qwen3.5 122B A10B is a open-weights model in the Qwen3.5 family from Alibaba. Benchmarked on 7 evals, averaging 59.7.
Benchmark scores · 7
avg 59.7 — higher is betterThe research behind this model
all papers mentioning it →It's Not the Capability: Harness Sensitivity Is Non-Monotone Across LLM Agent TiersMay 26, 2026XFP: Quality-Targeted Adaptive Codebook Quantization with Sparse Outlier Separation for LLM InferenceMay 14, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed147 tok/s
Latency (TTFT)1.1s
First answer15s
Price in / out$0.4 / $3.2 per 1M
Blended price$1.1 / 1M
Coding index45.7
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$3.2/M