MDL
Qwen3.5 35B A3B
Alibaba
Verified against Artificial Analysis · Jul 13, 2026
Capability
81
percentile index
Price /M
$0.25 / $2
in / out
Context
262.1K
tokens
Avg score
56.1
7 benchmarks
AA Intelligence
29.3
index
Speed
170 tok/s
TTFT 1.1s
Released
Feb 24, 2026
About
Qwen3.5 35B A3B is a open-weights model in the Qwen3.5 family from Alibaba. Benchmarked on 7 evals, averaging 56.1.
Benchmark scores · 7
avg 56.1 — higher is betterThe research behind this model
all papers mentioning it →Answer Presence Drives RAG Rewriting Gains▲ 8 on HF · Jun 4, 2026Diagnosing Overhead in Dispatch Operations: Cross-architecture ObservatoryMay 20, 2026Memory Grafting: Scaling Language Model Pre-training via Offline Conditional MemoryMay 20, 2026ClinSeekAgent: Automating Multimodal Evidence Seeking for Agentic Clinical ReasoningMay 19, 2026Negation Neglect: When models fail to learn negations in trainingMay 13, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed170 tok/s
Latency (TTFT)1.1s
First answer13s
Price in / out$0.25 / $2 per 1M
Blended price$0.688 / 1M
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$2/M