Model comparison
Kimi K3 vs Mercury 2
Kimi K3 is the stronger model overall, scoring 59.5 to 39.1 on the Noometry Index. Mercury 2 costs 16× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
Summary
- They share 16 benchmarks with published results for both. Kimi K3 scores higher in 7 categories and Mercury 2 in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 23.8.
- The biggest single-benchmark swing is WeirdML: 82.6% for Kimi K3 and 43.2% for Mercury 2.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 128K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Kimi K3 | Mercury 2 | |
|---|---|---|
| Provider | Moonshot AI | Inception |
| Noometry Index | 59.5 | 39.1 |
| Released | 2026-07-16 | 2026-02-20 |
| Weights | Open | Proprietary |
| Context window | 1.05M | 128K |
| Max output | 1.05M | 50K |
| Input $ / M tokens | $3 | $0.25 |
| Output $ / M tokens | $15 | $0.75 |
| Results tracked | 53 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Kimi K3: 61.0 (#10), Mercury 2: 33.5 (#255)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| LMArena WebDev | 1654 | 1171 |
| SciCode | 59.5% | 38.7% |
| WeirdML | 82.6% | 43.2% |
| LMArena Coding | 1508 | 1391 |
| ALE-Bench | 1,524 | 785.58 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| FrontierSWE | 25.9% | — |
Agentic & Tool Use Not comparable
Kimi K3: 41.8 (#20), Mercury 2: —
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| APEX-Agents | 50.6% | — |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 32% | — |
| GBAEval | 48.3% | — |
| GDP.pdf | 19% | — |
| Vending-Bench 2 | 5,165 | — |
Reasoning Kimi K3 leads
Kimi K3: 63.0 (#17), Mercury 2: 23.8 (#170)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| CritPt | 23.4% | 0.8% |
| LMArena Hard Prompts | 1496 | 1362 |
| ARC-AGI-2 | 60.4% | — |
| SimpleBench | 60.7% | — |
| NYT Connections (extended) | 93.6% | — |
| ARC-AGI-1 | 94.5% | — |
| Chess Puzzles | 39% | — |
| Mystery Game Puzzles | 26% | — |
| DTBench | 91.2% | — |
| LMCA | 52.7% | — |
| Surface Evolver Bench | 95% | — |
| Epoch Capabilities Index | 157.45 | — |
| ForecastBench | 61.1 | — |
Math Not comparable
Kimi K3: 74.2 (#16), Mercury 2: —
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 72.2% | — |
| FrontierMath Tier 4 | 39% | — |
| MathArena Final-Answer Competitions | 87.8% | — |
| OTIS Mock AIME 2024-2025 | 97.2% | — |
| ProofBench | 87% | — |
| LMArena Math | 1491 | — |
Knowledge Kimi K3 leads
Kimi K3: 63.2 (#21), Mercury 2: 36.2 (#172)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| LMArena Expert | 1521 | 1358 |
| GPQA Diamond | 93.1% | — |
| SimpleQA Verified | 50.6% | — |
| Vectara Hallucination Rate | — | 12.3% |
Multimodal Not comparable
Kimi K3: 37.8 (#70), Mercury 2: —
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |
Multilingual Kimi K3 leads
Kimi K3: 56.3 (#21), Mercury 2: 46.6 (#157)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1466 | 1331 |
| LMArena Chinese | 1529 | 1417 |
| LMArena Russian | 1482 | 1304 |
| LMArena French | 1491 | — |
| LMArena German | 1488 | — |
| LMArena Japanese | 1487 | — |
| LMArena Korean | 1458 | — |
| LMArena Spanish | 1472 | — |
Instruction Following Kimi K3 leads
Kimi K3: 77.7 (#14), Mercury 2: 70.2 (#165)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1483 | 1329 |
Long Context Kimi K3 leads
Kimi K3: 45.8 (#29), Mercury 2: 40.5 (#154)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1494 | 1330 |
Writing & Preference Kimi K3 leads
Kimi K3: 76.6 (#4), Mercury 2: 53.8 (#155)
| Benchmark | Kimi K3 | Mercury 2 |
|---|---|---|
| LMArena Text | 1476 | 1355 |
| LMArena Creative Writing | 1454 | 1289 |
| LMArena Multi-Turn | 1488 | 1358 |
| EQ-Bench Creative Writing | 2082 | — |
| EQ-Bench 4 | 1339 | — |
Frequently asked questions
Is Kimi K3 better than Mercury 2?
Kimi K3 is the stronger model overall, scoring 59.5 to 39.1 on the Noometry Index. Mercury 2 costs 16× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Kimi K3 or Mercury 2?
Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Kimi K3 lists at $3 and $15.
Is Kimi K3 or Mercury 2 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 128K.
How many benchmarks do Kimi K3 and Mercury 2 share?
16 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and Mercury 2 has 17.