MDL
Gemma 4 31B
Verified against Artificial Analysis · Jul 13, 2026
Capability
83
percentile index
Price /M
$0 / $0
in / out
Context
—
Avg score
55.1
7 benchmarks
AA Intelligence
29.4
index
Speed
35 tok/s
TTFT 0.9s
Released
Apr 2, 2026
About
Gemma 4 31B is a open-weights model in the Gemma family from Google. Benchmarked on 7 evals, averaging 55.1.
Benchmark scores · 7
avg 55.1 — higher is betterThe research behind this model
all papers mentioning it →Gemma 4 Technical Report▲ 60 on HF · Jul 2, 2026Discrete Diffusion Language Models for Interactive Radiology Report Drafting▲ 11 on HF · Jul 1, 2026RepSelect: Robust LLM Unlearning via Representation Selectivity▲ 4 on HF · Jun 15, 2026The Chain Holds, the Answer Folds: Trace-Answer Dissociation in Reasoning Models Under Adversarial Pressure▲ 1 on HF · May 27, 2026BioRefusalAudit: Auditing Biosecurity Refusal Depth Using General and Domain-Fine-Tuned Sparse AutoencodersMay 28, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed35 tok/s
Latency (TTFT)0.9s
First answer50s
Price in / out$0 / $0 per 1M
Coding index43.4
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$0/M