MDL
Qwen3 8B
Alibaba
Verified against Artificial Analysis · Jul 13, 2026
Capability
45
percentile index
Price /M
$0.18 / $2.1
in / out
Context
131.1K
tokens
Avg score
34.0
11 benchmarks
AA Intelligence
8.3
index
Speed
64 tok/s
TTFT 1.5s
Released
Apr 28, 2025
About
Qwen3 8B is a open-weights model in the Qwen3 family from Alibaba. Benchmarked on 11 evals, averaging 34.0.
Benchmark scores · 11
avg 34.0 — higher is betterThe research behind this model
all papers mentioning it →KVpop -- Key-Value Cache Compression with Predictive Online Pruning▲ 20 on HF · Jul 6, 2026Logit-Contribution Scoring Identifies Non-Literal Retrieval Heads▲ 18 on HF · Jul 1, 2026Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals▲ 12 on HF · May 25, 2026Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding▲ 11 on HF · Jul 7, 2026TREK: Distill to Explore, Reinforce to Refine▲ 8 on HF · Jul 6, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed64 tok/s
Latency (TTFT)1.5s
First answer33s
Price in / out$0.18 / $2.1 per 1M
Blended price$0.66 / 1M
Coding index9.0
Math index19.0
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$2.1/M