vs
0:0
overall capability wins (1 ties across 1 scored). Mercury 2 leads overall.
- Pricing0:3
Verdict
Aya Expanse 32B ships open weights (CC-BY-NC-4.0); Mercury 2 is ~2.0x cheaper on a blended token basis than Aya Expanse 32B.
- Aya Expanse 32B ships open weights (CC-BY-NC-4.0)
- Mercury 2 is ~2.0x cheaper on a blended token basis than Aya Expanse 32B
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Aya Expanse 32B
- $0.88
- Mercury 2
- $0.44
Mercury 2 is about 2.0x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Cohere
Inception Labs
Family
Other
Other
License
CC-BY-NC-4.0
Proprietary
Open weights
Yes
No
Release
Nov 1, 2024
Sep 1, 2026
Knowledge cutoff
-
-
API / provider
Cohere
Inception Labs
Modalities
text → text
text → text
Specs
Context window
128K
128K
Max output
4K
-
Parameters
32B
—
Pricing
Input $/1M
$0.50
$0.25
Output $/1M
$1.50
$0.75
Blended $/1M (3∶1)
$0.75
$0.38
Speed
tok/s
85
-
TTFT (s)
0.3
-
Reasoning
MMLU-Pro
—
—
GPQA Diamond
—
77.0
Humanity's Last Exam
—
17.1
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
—
21.2
AA-LCR v1.1
—
43.7
CritPt
—
0.8
MMMU-Pro
—
—
IFBench
—
69.8
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
—
27.3
Aider Polyglot
—
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
—
37.7
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
1,167
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
—
—
| Metric | Aya Expanse 32B | Mercury 2 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Cohere | Inception Labs | — |
| Family | Other | Other | — |
| License | CC-BY-NC-4.0 | Proprietary | — |
| Open weights | Yes | No | — |
| Release | Nov 1, 2024 | Sep 1, 2026 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Cohere | Inception Labs | — |
| Modalities | text → text | text → text | — |
| Specs | |||
| Context window | 128K | 128K | tie |
| Max output | 4K | - | — |
| Parameters | 32B | — | — |
| Pricing | |||
| Input $/1M | $0.50 | $0.25 | B +$0.25 |
| Output $/1M | $1.50 | $0.75 | B +$0.75 |
| Blended $/1M (3∶1) | $0.75 | $0.38 | B +$0.38 |
| Speed | |||
| tok/s | 85 | - | — |
| TTFT (s) | 0.3 | - | — |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | — | 77.0 | — |
| Humanity's Last Exam | — | 17.1 | — |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | — | 21.2 | — |
| AA-LCR v1.1 | — | 43.7 | — |
| CritPt | — | 0.8 | — |
| MMMU-Pro | — | — | — |
| IFBench | — | 69.8 | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | — | 27.3 | — |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | — | 37.7 | — |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | 1,167 | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | — | — | — |
Aya Expanse 32B
- Context
- 128K / 4K out
- Parameters
- 32B
- Price
- $0.50 / $1.50
- Speed
- 85 tok/s · 0.3s TTFT
- Modalities
- text → text
- License
- CC-BY-NC-4.0
C4AI research multilingual model covering 23 languages — strong open alternative for global chat and translation.
Mercury 2
- Context
- 128K
- Parameters
- —
- Price
- $0.25 / $0.75
- Speed
- —
- Modalities
- text → text
- License
- Proprietary
Inception Labs' diffusion-based (non-autoregressive) reasoning LLM, API-only with a 128K context window and high-throughput inference on Blackwell GPUs.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.
These two models have no overlapping published benchmarks in our dataset. Compare specs and pricing instead.