vs
1:1
overall capability wins (0 ties across 2 scored). Price/speed category wins do not decide overall. Tied overall.
- Specs1:1
Verdict
K2 Horizon 375B A23B offers ~128.0x larger context (524K vs 4K).
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- K2 Horizon 375B A23B
- -
- Snowflake Arctic Instruct
- $3
Price per 1M tokens
Full comparison
Organization
MBZUAI Institute of Foundation Models
Snowflake
Family
Other
Other
License
Apache 2.0
Apache 2.0
Open weights
Yes
Yes
Release
Sep 3, 2026
Apr 24, 2024
Knowledge cutoff
-
-
API / provider
No primary API price listed
Together / Fireworks (ref.)
Modalities
text → text
text → text
Specs
Context window
524K
4K
Max output
-
4K
Parameters
375B (23B act.)
480B (17B act.)
Pricing
Input $/1M
—
$2.40
Output $/1M
—
$2.40
Blended $/1M (3∶1)
-
$2.40
Speed
tok/s
-
45
TTFT (s)
-
0.55
Reasoning
MMLU-Pro
—
—
GPQA Diamond
87.3
—
Humanity's Last Exam
32.0
—
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
18.2
—
AA-LCR v1.1
80.0
—
CritPt
4.6
—
MMMU-Pro
—
—
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
71.9
—
Aider Polyglot
—
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
42.9
—
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
—
1,180
| Metric | K2 Horizon 375B A23B | Snowflake Arctic Instruct | Delta |
|---|---|---|---|
| Identity | |||
| Organization | MBZUAI Institute of Foundation Models | Snowflake | — |
| Family | Other | Other | — |
| License | Apache 2.0 | Apache 2.0 | — |
| Open weights | Yes | Yes | — |
| Release | Sep 3, 2026 | Apr 24, 2024 | — |
| Knowledge cutoff | - | - | — |
| API / provider | No primary API price listed | Together / Fireworks (ref.) | — |
| Modalities | text → text | text → text | — |
| Specs | |||
| Context window | 524K | 4K | A +520K |
| Max output | - | 4K | — |
| Parameters | 375B (23B act.) | 480B (17B act.) | B +105 |
| Pricing | |||
| Input $/1M | — | $2.40 | — |
| Output $/1M | — | $2.40 | — |
| Blended $/1M (3∶1) | - | $2.40 | — |
| Speed | |||
| tok/s | - | 45 | — |
| TTFT (s) | - | 0.55 | — |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | 87.3 | — | — |
| Humanity's Last Exam | 32.0 | — | — |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 18.2 | — | — |
| AA-LCR v1.1 | 80.0 | — | — |
| CritPt | 4.6 | — | — |
| MMMU-Pro | — | — | — |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | 71.9 | — | — |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | 42.9 | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | — | 1,180 | — |
K2 Horizon 375B A23B
- Context
- 524K
- Parameters
- 375B (23B act.)
- Price
- —
- Speed
- —
- Modalities
- text → text
- License
- Apache 2.0
Flagship model (375B MoE / 23B active) in the Institute of Foundation Models' K2 Horizon fleet, released fully open — weights, code, training data, and methodology — under Apache 2.0.
Snowflake Arctic Instruct
- Context
- 4K / 4K out
- Parameters
- 480B (17B act.)
- Price
- $2.40 / $2.40
- Speed
- 45 tok/s · 0.55s TTFT
- Modalities
- text → text
- License
- Apache 2.0
480B dense-MoE hybrid tuned for enterprise SQL, coding, and instruction following — Apache 2.0 weights from Snowflake AI Research.
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.