whatmodel.sh / Nova 2 Lite
Nova 2 Lite
↗ share
+ contribute
Nova 2 Lite is a closed-weights language model from Amazon, released October 2025.
$ how it compares
The LLMs ranked either side of Nova 2 Lite, out of 828 with a score, then more from Amazon.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 267 |  Qwen3 32B | 58±0 | api $0.13/M +local | ||
| 268 | Qwen3 Coder 30B-A3B Instruct | 58±1 | api $0.12/M +local | ||
| 269 | Nova 2 Lite | 58±2 | api $0.94/M | ||
| 270 |  Llama 3.3 Nemotron Super 49B v1.5 | 58±1 | local | ||
| 271 | Nemotron 3 Nano 30B-A3B | 58±1 | api $0.087/M +local | ||
| more from Amazon | |||||
| 184 | Nova 2 Pro (Preview) | 64±2 | api $3.78/M | ||
| 339 | Nova Premier | 54±2 | api $5.00/M | ||
| 423 | Nova Pro | 50±1 | api $1.40/M | ||
$ what it's good at
agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #229 of 300 * · 15% of its score · 2 tests
Beats o3-mini here, though they rank higher overall. Trails SmolLM3 3B, Phi-3.5-mini-instruct, Jamba 1.5 Mini here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| BFCL V4 (December 2025 archive) | 27.1% ✎ | ▁▁▁▂██▅▅▆▂▂▂ beats 21/77 | 10% |
| τ²-Bench-Verified (GPT-5.1 user) lab | 72.4% ✎ | ▂··▄···▂▄▄█▄ beats 4/13 | 4.9% |
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #175 of 393 * · 21% of its score · 1 test
Beats Seed-OSS 36B Instruct, Nemotron-Cascade-14B-Thinking, Qwen3-VL 32B Instruct here, though they rank higher overall. Trails LFM2.5 1.2B Instruct, LFM2.5 1.2B Thinking, Gemma 3 12B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena (text) | 1335.0 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 170/379 | 21% |
other tests4 not in a task · 63% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| CLIcK Korean (Daekeun Kim) | 67.2% ✎ | ▆▄▄▄█▄▂·▄█·▄ beats 7/23 | 26% |
| HAE-RAE 1.0 (Daekeun Kim) | 73.9% ✎ | ▄▄▄▂▃█▂▄▂█▂▄ beats 7/22 | 21% |
| KMMLU-HARD (Daekeun Kim, zero-shot) | 48.2% ✎ | █···▁▁▁▃▃·▆▂ beats 12/23 | 15% |
| Vectara hallucination leaderboard (HHEM-2.3, factual consistency rate) | 94.9% ✎ | ▁▂▁▂▅█▂▃▄▃▁▁ beats 90/98 | 0.0% |
$ where to run it
Through an API
api · Amazon Bedrock · $0.94/M · $0.33 in / $2.75 out per 1M tokens · 1M context · docs.aws.amazon.com ↗ · price sourceUSD per 1M tokens, Standard tier, geo cross-Region inference and in-Region, US East (N. Virginia); global cross-Region inference $0.30/$2.50. Observed 2026-09-25.
$ how the score was worked out · Daekeun Kim Korean LLM benchm… 63%, LMArena leaderboard dataset 21%, Berkeley Function Calling Lea… 10% and 2 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Daekeun Kim Korean LLM benchmark runs | 3 | 63% |
 LMArena leaderboard dataset | 1 | 21% |
| Berkeley Function Calling Leaderboard result archive | 1 | 10% |
| Amazon AGI — τ²-Bench-Verified · lab-reported | 1 | 5% |
 Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository) | 1 | 0% |
| prior · pull toward the middle | 1% | |
| weighted average | = 57.5 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json