nemotron-3-nano:4b
Family: nemotron-3-nano · 4B parameters · 1 run(s)
Fits in 8 GB VRAM · ollama 0.40.6 · Q4_K_M
Measured VRAM at 32K context (KV cache included, ollama ps on the RTX 3090): 3.5 GB. Tier rule
79.0/100 · scoring v4
95% task-bootstrap CI: 73.6–84.0 · rank #4 in tier, rank range under weight sensitivity #4–#4 (5th–95th percentile; min #4, max #5) · v3: 78.1 · mention 4.4% (2/45) · how
Plain-language summary
Solid for agents — handles most tool tasks well.
Good for
- returning clean JSON
- matching the requested format
- picking the right tool
Weak at
- —
Category breakdown
| JSON | Schema | Tool | Multi-step | Injection (useful&safe) | tok/s | p50 s/task | p95 s/task |
|---|---|---|---|---|---|---|---|
| 100 | 91 | 96 | 61 | 64 | — | 12.2 | 20.2 |
Prompt injection (15 tasks × 3 trials): useful and safe 64.4% (task completed and not compromised; counts in the score) · utility under attack —% (completed the user's task) · harmful-action attack success 0.0% (a real harmful tool call or canary leak). Mentions (quoting the attacker without acting): 4.4% (2/45); they do not lower the score. v3 useful and safe: 60.0%. v2.1 used injection resistance 95.6 × tool-call rate . Definitions
- Run:
20261112-nemotron-3-nano-4b-1dde07· 2026-11-12T06:25:37+00:00 - GPU: NVIDIA GeForce RTX 3090 (24 GB) · Linux 6.8.0-57-generic
- Host:
· kind not recorded · measured · hosts - Harness: hermes-agent Hermes Agent v0.22.2+2363.g9cb34e9 (2026.9.24) · stable · local 9cb34e9e285d · trials 3 · sampling: Ollama default (temperature and seed not set by harness) · context window: 32768 tokens as loaded by Ollama (the unused runner field records 8192; see Methodology)
- GGUF sha256:
527db2cf6c705d8fabb95693d038d9c06b4a2b0b8b0a4bbdbd01212d37242970 - Sponsor disclosure: none
Chat template vs upstream
No reference
Labels are only “matches / differs / no reference”, never safe/unsafe. Ollama may use a built-in Go template or renderer instead of the GGUF Jinja. Why templates matter: Chat-template backdoors. Not part of the score.