Model comparison
Kimi K2.5 vs Nvidia Llama 3.3 Nemotron Super 49b v1.5
Kimi K2.5 is the stronger model overall, scoring 48.1 to 40.3 on the Noometry Index. Nvidia Llama 3.3 Nemotron Super 49b v1.5 costs 2.3× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
NVIDIASummary
- They share 12 benchmarks with published results for both. Kimi K2.5 scores higher in 8 categories and Nvidia Llama 3.3 Nemotron Super 49b v1.5 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Kimi K2.5 leads 53.6 to 36.7.
- Nvidia Llama 3.3 Nemotron Super 49b v1.5 is cheaper at $0.40 / $0.40 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- Kimi K2.5 accepts more context: 262K tokens versus 131K.
Side by side
| Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 | |
|---|---|---|
| Provider | Moonshot AI | NVIDIA |
| Noometry Index | 48.1 | 40.3 |
| Released | 2026-01-27 | 2025-07-25 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 262K | 131K |
| Input $ / M tokens | $0.45 | $0.40 |
| Output $ / M tokens | $2.25 | $0.40 |
| Results tracked | 51 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Kimi K2.5: 48.8 (#53), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 39.8 (#154)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Coding | 1474 | 1355 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| LMArena WebDev | 1437 | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| WeirdML | 45.6% | — |
| ALE-Bench | 821.65 | — |
Agentic & Tool Use Not comparable
Kimi K2.5: 34.2 (#48), Nvidia Llama 3.3 Nemotron Super 49b v1.5: —
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| OSWorld | 63.3% | — |
| Vending-Bench 2 | 1,198 | — |
Reasoning Kimi K2.5 leads
Kimi K2.5: 31.2 (#80), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 26.8 (#128)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Hard Prompts | 1453 | 1336 |
| ARC-AGI-2 | 11.8% | — |
| SimpleBench | 46.8% | — |
| Kagi LLM Benchmark | 78.5% | — |
| NYT Connections (extended) | 69.9% | — |
| ARC-AGI-1 | 65.3% | — |
| CritPt | 3.1% | — |
| Chess Puzzles | 12% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| Epoch Capabilities Index | 148.03 | — |
Math Kimi K2.5 leads
Kimi K2.5: 51.8 (#53), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 38.2 (#141)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Math | 1470 | 1392 |
| MathArena Final-Answer Competitions | 62.3% | — |
| OTIS Mock AIME 2024-2025 | 92.2% | — |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K2.5 leads
Kimi K2.5: 53.6 (#56), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 36.7 (#165)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Expert | 1466 | 1330 |
| GPQA Diamond | 87.6% | — |
| Humanity's Last Exam | 24.4% | — |
| SimpleQA Verified | 34.3% | — |
| Vectara Hallucination Rate | 14.2% | — |
Multimodal Not comparable
Kimi K2.5: 41.1 (#39), Nvidia Llama 3.3 Nemotron Super 49b v1.5: —
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |
Multilingual Kimi K2.5 leads
Kimi K2.5: 53.9 (#53), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 45.5 (#168)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Non-English | 1433 | 1316 |
| LMArena Japanese | 1421 | 1300 |
| LMArena Russian | 1435 | 1332 |
| LMArena Chinese | 1495 | — |
| LMArena French | 1454 | — |
| LMArena German | 1441 | — |
| LMArena Korean | 1410 | — |
| LMArena Spanish | 1450 | — |
Instruction Following Kimi K2.5 leads
Kimi K2.5: 75.3 (#64), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 68.6 (#188)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Instruction Following | 1431 | 1299 |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 40.0 (#164)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Longer Query | 1445 | 1315 |
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.1 (#53), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 53.1 (#159)
| Benchmark | Kimi K2.5 | Nvidia Llama 3.3 Nemotron Super 49b v1.5 |
|---|---|---|
| LMArena Text | 1445 | 1338 |
| LMArena Creative Writing | 1423 | 1307 |
| LMArena Multi-Turn | 1444 | 1334 |
| EQ-Bench Creative Writing | 1579 | — |
Frequently asked questions
Is Kimi K2.5 better than Nvidia Llama 3.3 Nemotron Super 49b v1.5?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 40.3 on the Noometry Index. Nvidia Llama 3.3 Nemotron Super 49b v1.5 costs 2.3× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Which is cheaper, Kimi K2.5 or Nvidia Llama 3.3 Nemotron Super 49b v1.5?
Nvidia Llama 3.3 Nemotron Super 49b v1.5 is cheaper. It lists at $0.40 per million input tokens and $0.40 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is Kimi K2.5 or Nvidia Llama 3.3 Nemotron Super 49b v1.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 39.8 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 131K.
How many benchmarks do Kimi K2.5 and Nvidia Llama 3.3 Nemotron Super 49b v1.5 share?
12 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Nvidia Llama 3.3 Nemotron Super 49b v1.5 has 12.