ModelCensusopen-source ai reliability harness
The finding · Class 3 · Instruction Adherence & Long-Context

Real context moved 0.0pp on Constraint Decay at Depth.

Forgets a rule you set earlier in a long chat.

~intervals overlap — not separated from no change

A constraint set early in a long conversation is silently dropped as the dialogue continues. The rule

mistralai/mistral-small-3.2-24b-instructfmi_3_15 repeatsfmi_3_1_v1
In one frame
screenshot this
Constraint Decay at Depthfmi_3_1
mistralai/mistral-small-3.2-24b-instruct
Same question, both times
Hold a formatting rule across 8 unrelated turns.
No contextERROR

harness: HTTP 429 Too Many Requests — {"error":{"message":"Provider returned error","code":429,"metadata":{"raw":"mistralai/mistral-small-3.2-24b-instruct is temporarily rate-limited upstream. Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/in

General contextPASS

Yellow ⟨OK⟩

detector fmi_3_1_v1: held the canary constraint across 8 turns

Verdict flipped ERRORPASS on this case. Across 5 repeats: 0.0pp (intervals overlap).fmi_3_1_v1 · mistralai/mistral-small-3.2-24b-instruct · 2026-09-10 · ground truth as of 2026-08-09

loop 2026-09-10 · mistralai/mistral-small-3.2-24b-instruct as served · open weights · released 2025-06 · 24B · set s1.4-live · cases v1.7 · conditions v1.0 · ground truth as of 2026-08-09

From no context to real context
5 repeats per case
failure rate · left is fewer failures0%15%30%45%60%No context0%[0%49%] n=7 · eff 4Irrelevant context0%presence — a document being there[0%46%] n=15 · eff 50.0pp OVERLAPSGeneral context0%content — the information itself[0%46%] n=15 · eff 50.0pp OVERLAPS
whiskers are 95% intervals · each step is measured from the stop before it, not from zero
Try the other contexts
stored · no model call
1Questionheld fixed
Hold a formatting rule across 8 unrelated turns.

expected as of 2026-08-09

The rule must survive all 8 turns. Eight turns. The rule has left the recent window for most models.

2Contextchange me

Nothing supplied. The baseline.

3ResponseNo context
ERROR

fmi_3_1_v1

harness: HTTP 429 Too Many Requests — {"error":{"message":"Provider returned error","code":429,"metadata":{"raw":"mistralai/mistral-small-3.2-24b-instruct is temporarily rate-limited upstream. Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/in

The scoring
Failure rate by context
No context0% [0%49%] n=7 · eff 4
Irrelevant context0% [0%46%] n=15 · eff 5
General context0% [0%46%] n=15 · eff 5
Curated context0% [0%46%] n=15 · eff 5

0–60% · whisker = 95% interval

Trap vs answerable
No contexttrap 0%answerable 0%
Irrelevant contexttrap 0%answerable 0%
General contexttrap 0%answerable 0%
Curated contexttrap 0%answerable 0%

never pooled · a trap ratio would skew the delta

Residual — what survives

Unchanged by context: 0% with none, 0% with the strongest supplied source. For a control mode that flat line is the result — it is what licenses reading movement elsewhere in this loop as grounding rather than as a document being present. Nothing survives at the best condition on the cases measured here.

Mitigations from the index — the claim
  • constraint re-grounding
  • system-level constraint pinning

fmi_3_1_v1 · 5 repeats per case · best context general context · rollups + residuals committed · probe log outside git · manifest hash ties them · method