Devstral 2 (2512) devstral-2-2512

mistralai · mistral coding agentic

unranked — no quality signals yet

params
123B
arch
dense
context
256k
license
other open weights
released
Nov 2025
downloads/30d
19.6k

# architecture

attn shared GQA 96q/8kv feed-forward (dense) every parameter, every token out ×88 layers ctx 262,144 123B — no routing
layers
88
d_model
12288
heads
96q / 8kv
head dim
128
vocab
131072
family
ministral3

signal path of one layer, generated from the registry's structured fields — dimension lines quote the model's real numbers; an MoE trace forks at the router, a dense trace runs straight through. Geometry fields come from the repo's config.json.

# why ranked

Overall: provisional — score 35.9 ● low — a single signal; treat with caution

provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.

signalweightinput (0–100)
aa_intelligence 0.70
bench_composite 0.30 35.9

benchmark panel evidence:

complete panelscore (0–100)
swe-bench-verified-v135.9

missing signals are dropped and the remaining weights renormalized — never imputed.

Coding: provisional — score 22.2 ● low — a single signal; treat with caution

provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.

signalweightinput (0–100)
aa_coding 0.30
aider_polyglot 0.30
swe_bench_verified 0.40 22.2

missing signals are dropped and the remaining weights renormalized — never imputed.

Agentic: #17 — score 22.2 ● low — a single signal; treat with caution

signalweightinput (0–100)
swe_bench_verified 0.60 22.2
swe_rebench 0.40

missing signals are dropped and the remaining weights renormalized — never imputed.

full methodology

# trend

OGM score · last 54 days
OGM score over 54 days

methodology changed during this history window; score movement across that boundary is not model movement. See methodology v5.

# benchmarks

benchmarkscoresourcedate
SWE-bench Verified 53.8 SWE-bench Dec 2025

# where to run

providerquantctx$/M in$/M out$/M cacheprice srctpsuptime
Mistralunknown 256k $0.40$2.00$0.04 via openrouter 100.0%
Scaleway 200k $0.40$2.00 via litellm

sorted by blended price ((3·input + output) / 4 per 1M) · ✓ = the provider's own catalog confirms the offer · "via …" prices are what the aggregator routing the offer charges, not the provider's own list price

source aliases
mistral
devstral-2512
openrouter
mistralai/devstral-2512
swebench
devstral-2512