MDL
Qwen3 235B A22B 2507
Alibaba
Verified against Artificial Analysis · Jul 13, 2026
Capability
81
percentile index
Price /M
$0.7 / $8.4
in / out
Context
262.1K
tokens
Avg score
61.3
11 benchmarks
AA Intelligence
19.6
index
Speed
68 tok/s
TTFT 1.3s
Released
Jul 25, 2025
About
Qwen3 235B A22B 2507 is a open-weights model in the Qwen3 family from Alibaba. Benchmarked on 11 evals, averaging 61.3.
Benchmark scores · 11
avg 61.3 — higher is betterThe research behind this model
all papers mentioning it →ACC: Compiling Agent Trajectories for Long-Context TrainingMay 21, 2026Draft Less, Retrieve More: Hybrid Tree Construction for Speculative DecodingMay 19, 2026GoLongRL: Capability-Oriented Long Context Reinforcement Learning with Multitask AlignmentMay 19, 2026FORGE: Self-Evolving Agent Memory With No Weight Updates via Population BroadcastMay 15, 2026Persona-Model Collapse in Emergent MisalignmentMay 13, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed68 tok/s
Latency (TTFT)1.3s
First answer31s
Price in / out$0.7 / $8.4 per 1M
Blended price$2.63 / 1M
Coding index22.1
Math index91.0
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$8.4/M