Model comparison
Kimi K2 (Jul 2025) vs Mercury 2
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 39.1 on the Noometry Index. Mercury 2 costs 2.7× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
Summary
- They share 14 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 6 categories and Mercury 2 in 1 category; 4 gaps are clear of the uncertainty.
- The widest gap is in coding, where Kimi K2 (Jul 2025) leads 42.4 to 33.5.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 17.9% for Kimi K2 (Jul 2025) and 12.3% for Mercury 2.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 128K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Kimi K2 (Jul 2025) | Mercury 2 | |
|---|---|---|
| Provider | Moonshot AI | Inception |
| Noometry Index | 41.2 | 39.1 |
| Released | 2025-07-12 | 2026-02-20 |
| Weights | Open | Proprietary |
| Context window | 262K | 128K |
| Max output | 262K | 50K |
| Input $ / M tokens | $0.57 | $0.25 |
| Output $ / M tokens | $2.30 | $0.75 |
| Results tracked | 42 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.4 (#102), Mercury 2: 33.5 (#255)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| WeirdML | 42.8% | 43.2% |
| LMArena Coding | 1399 | 1391 |
| ALE-Bench | 597.5 | 785.58 |
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| LMArena WebDev | — | 1171 |
| SciCode | — | 38.7% |
| GSO | 4.9% | — |
Agentic & Tool Use Not comparable
Kimi K2 (Jul 2025): 32.4 (#64), Mercury 2: —
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| METR Time Horizons | 59.2% | — |
Reasoning Too close to call
Kimi K2 (Jul 2025): 23.3 (#179), Mercury 2: 23.8 (#170)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| LMArena Hard Prompts | 1384 | 1362 |
| SimpleBench | 26.3% | — |
| Kagi LLM Benchmark | 64.4% | — |
| CritPt | — | 0.8% |
| Epoch Capabilities Index | 146.01 | — |
| ForecastBench | 60.2 | — |
Math Not comparable
Kimi K2 (Jul 2025): 42.7 (#83), Mercury 2: —
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| Omni-MATH | 65.4% | — |
| LMArena Math | 1397 | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 37.3 (#157), Mercury 2: 36.2 (#172)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| Vectara Hallucination Rate | 17.9% | 12.3% |
| LMArena Expert | 1365 | 1358 |
| MMLU-Pro | 81.9% | — |
| Confabulations | 20.4% | — |
| GPQA (HELM) | 65.3% | — |
Multilingual Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 49.6 (#130), Mercury 2: 46.6 (#157)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1372 | 1331 |
| LMArena Chinese | 1415 | 1417 |
| LMArena Russian | 1385 | 1304 |
| LMArena French | 1379 | — |
| LMArena German | 1387 | — |
| LMArena Japanese | 1349 | — |
| LMArena Korean | 1325 | — |
| LMArena Spanish | 1386 | — |
Instruction Following Too close to call
Kimi K2 (Jul 2025): 71.1 (#156), Mercury 2: 70.2 (#165)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1348 | 1329 |
| IFEval | 85% | — |
Long Context Too close to call
Kimi K2 (Jul 2025): 41.2 (#145), Mercury 2: 40.5 (#154)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1353 | 1330 |
| Fiction.LiveBench | 66.7% | — |
| CL-bench | 17.6% | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), Mercury 2: 53.8 (#155)
| Benchmark | Kimi K2 (Jul 2025) | Mercury 2 |
|---|---|---|
| LMArena Text | 1380 | 1355 |
| LMArena Creative Writing | 1350 | 1289 |
| LMArena Multi-Turn | 1371 | 1358 |
| Short-Story Creative Writing | 85.6% | — |
| EQ-Bench Creative Writing | 1666 | — |
| WildBench | 86.2% | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Mercury 2?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 39.1 on the Noometry Index. Mercury 2 costs 2.7× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Which is cheaper, Kimi K2 (Jul 2025) or Mercury 2?
Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is Kimi K2 (Jul 2025) or Mercury 2 better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 128K.
How many benchmarks do Kimi K2 (Jul 2025) and Mercury 2 share?
14 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Mercury 2 has 17.