Mercury
Launch tracker · day 1
Mercury at launch: the first independent results
As of 19 Nov 2026, Mercury (Inception, first listed 19 Nov 2026) has independent results from LMArena (Arena Text 1305) and Kagi LLM Benchmark (Kagi LLM Benchmark 21.6%). It has no BenchLeader Index yet: the Index needs at least five independent results across two categories. Not yet measured by Artificial Analysis, Epoch AI, Vals AI, LiveBench, ARC Prize, Scale AI and SWE-bench.
- First listed
- 19 Nov 2026 (day 1)
- Provider
- Inception
- Price
- Not listed yet
- Context window
- Not listed yet
- Open weights
- No
- BenchLeader Index
- Not yet (needs five results in two categories)
| Independent board | Results so far |
|---|---|
| LMArena | Arena Text 1305 · Arena Hard Prompts 1318 · Arena Coding 1363 · more below |
| Kagi LLM Benchmark | Kagi LLM Benchmark 21.6% |
Boards usually post results within days of a launch; this page is rebuilt every three hours.
Scores from independent publishers only; benchmark figures a provider publishes about its own model are not shown.
As of 19 Nov 2026, Mercury by Inception is not yet ranked on the BenchLeader Index, which needs at least five independent results across two categories.
In detail. Mercury is an Inception proprietary model. It is not yet ranked: the index needs at least five independent results across two categories. Last measured 18 Nov 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- –
- Cost per task
- –
- Tokens per task
- –
- Released
- –
- date not published by our sources
- Reasoning34
- Coding48
- Human preference46
Versions
Inception has shipped 4 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Mercury 3high | 19 Oct 2026 | – | – |
| Mercury 2.5 | 8 Sep 2026 | – | – |
| Mercury 2 | 20 Feb 2026 | 49.2 | #378 |
| NewMercurythis page | – | – | – |
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source |
|---|---|---|
| Arena Hard Prompts | 1318#259 | Arena (LMArena) |
| Kagi LLM Benchmark | 21.6%#130 | Kagi LLM Benchmark |
Coding
| Benchmark | not stated | Source |
|---|---|---|
| Arena Coding | 1363#240 | Arena (LMArena) |
Instruction following
| Benchmark | not stated | Source |
|---|---|---|
| Arena Instruction Followingnot in index | 1270#279 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source |
|---|---|---|
| Arena Text | 1305#264 | Arena (LMArena) |
| Arena Creative Writingnot in index | 1218#296 | Arena (LMArena) |
| Arena Multi-turnnot in index | 1298#255 | Arena (LMArena) |
| Arena Longer Queriesnot in index | 1296#271 | Arena (LMArena) |
More from Inception
See also
Questions people ask
- Why does Mercury have no BenchLeader Index score?
- The index needs at least five independent results across at least two categories, and as of 19 Nov 2026 Mercury does not have them yet. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
Data as of 19 Nov 2026. Compare with another model.
Cite as: BenchLeader, “Mercury: benchmarks, pricing, speed and rank”, data as of 19 Nov 2026, https://www.benchleader.com/models/mercury.