Model comparison
Kimi K2.5 vs Llama 3.2 1B
Kimi K2.5 is the stronger model overall, scoring 47.8 to 19.9 on the Noometry Index. Llama 3.2 1B costs 13× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
MetaSummary
- They share 18 benchmarks with published results for both. Kimi K2.5 scores higher in 9 categories and Llama 3.2 1B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Kimi K2.5 leads 53.6 to 7.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Kimi K2.5 and 0.6% for Llama 3.2 1B.
- Llama 3.2 1B is cheaper at $0.027 / $0.20 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- Kimi K2.5 accepts more context: 262K tokens versus 60K.
Side by side
| Kimi K2.5 | Llama 3.2 1B | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 47.8 | 19.9 |
| Released | 2026-01-27 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 262K | 60K |
| Max output | 262K | 54K |
| Input $ / M tokens | $0.45 | $0.027 |
| Output $ / M tokens | $2.25 | $0.20 |
| Results tracked | 51 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Kimi K2.5: 48.7 (#48), Llama 3.2 1B: 20.6 (#333)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| LMArena Coding | 1474 | 1071 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| LMArena WebDev | 1436 | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| WeirdML | 45.6% | — |
| BigCodeBench Instruct | — | 8.2% |
| BigCodeBench Complete | — | 11.3% |
| ALE-Bench | 821.65 | — |
Agentic & Tool Use Kimi K2.5 leads
Kimi K2.5: 34.0 (#44), Llama 3.2 1B: 14.5 (#142)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| Berkeley Function Calling Leaderboard | — | 10.8% |
| OSWorld | 63.3% | — |
| BALROG | — | 6.6% |
| Vending-Bench 2 | 1,198 | — |
Reasoning Kimi K2.5 leads
Kimi K2.5: 31.2 (#74), Llama 3.2 1B: 15.6 (#304)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| Chess Puzzles | 12% | 0% |
| LMArena Hard Prompts | 1454 | 1044 |
| Epoch Capabilities Index | 148.05 | 101.95 |
| 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% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
Math Kimi K2.5 leads
Kimi K2.5: 50.3 (#46), Llama 3.2 1B: 9.7 (#308)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 0.6% |
| LMArena Math | 1470 | 1086 |
| MathArena Final-Answer Competitions | 62.3% | — |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K2.5 leads
Kimi K2.5: 53.6 (#52), Llama 3.2 1B: 7.2 (#307)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| GPQA Diamond | 87.6% | 23.9% |
| LMArena Expert | 1466 | 1007 |
| Humanity's Last Exam | 24.4% | — |
| SimpleQA Verified | 34.3% | — |
| Vectara Hallucination Rate | 14.2% | — |
Multimodal Not comparable
Kimi K2.5: 40.4 (#37), Llama 3.2 1B: —
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |
Multilingual Kimi K2.5 leads
Kimi K2.5: 54.1 (#50), Llama 3.2 1B: 23.9 (#287)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| LMArena Non-English | 1433 | 973 |
| LMArena Chinese | 1495 | 958 |
| LMArena German | 1441 | 1014 |
| LMArena Russian | 1436 | 941 |
| LMArena French | 1454 | — |
| LMArena Japanese | 1422 | — |
| LMArena Korean | 1412 | — |
| LMArena Spanish | 1450 | — |
Instruction Following Kimi K2.5 leads
Kimi K2.5: 75.5 (#60), Llama 3.2 1B: 52.6 (#285)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1031 |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Llama 3.2 1B: 31.9 (#269)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| LMArena Longer Query | 1445 | 1050 |
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.3 (#51), Llama 3.2 1B: 21.4 (#305)
| Benchmark | Kimi K2.5 | Llama 3.2 1B |
|---|---|---|
| LMArena Text | 1445 | 1055 |
| LMArena Creative Writing | 1423 | 1033 |
| EQ-Bench Creative Writing | 1579 | 200 |
| LMArena Multi-Turn | 1444 | 1030 |
Frequently asked questions
Is Kimi K2.5 better than Llama 3.2 1B?
Kimi K2.5 is the stronger model overall, scoring 47.8 to 19.9 on the Noometry Index. Llama 3.2 1B costs 13× 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 Llama 3.2 1B?
Llama 3.2 1B is cheaper. It lists at $0.027 per million input tokens and $0.20 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is Kimi K2.5 or Llama 3.2 1B better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.7 versus 20.6 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 60K.
How many benchmarks do Kimi K2.5 and Llama 3.2 1B share?
18 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Llama 3.2 1B has 22.