Language modelOpen weights
K2 Horizon 0.9B
Institute of Foundation Models
- Released
- September 3, 2026
- Data date
- October 7, 2026
- Input
- Text
- Output
- Text
K2 Horizon 0.9B is the family’s compact dense model. It supports up to 131,072 tokens with YaRN RoPE scaling and uses an additional on-policy distillation stage, which IFM calls MOPD, after pretraining, midtraining, and RL.
Specifications and access
| Specification | Value and source |
|---|---|
| Parameters | 0.9B total parametersSource |
| Active parameters | 0.9B parameters per tokenSource |
| Context window | 131,072 tokensSource |
| Input modalities | TextSource |
| Output modalities | TextSource |
| Documented training stages | Pretraining, two midtraining stages, RL expert models with a merge, and subsequent MOPD on-policy distillation.Source |
| License | Apache 2.0Source |
Page 1 of 2
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
HumanEval+ (Show measurement, test conditions, and source)
- Source value
- 79.9
- Score
- 79.9
- Metric
- HumanEval+
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
MBPP+ (Show measurement, test conditions, and source)
- Source value
- 68
- Score
- 68
- Metric
- MBPP+
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
BFCL v4 (Show measurement, test conditions, and source)
- Source value
- 28
- Score
- 28
- Metric
- BFCL v4
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
AIME 2025 (Show measurement, test conditions, and source)
- Source value
- 41.7
- Score
- 41.7
- Metric
- AIME 2025
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
AIME 2026 (Show measurement, test conditions, and source)
- Source value
- 48.5
- Score
- 48.5
- Metric
- AIME 2026
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
HMMT February 2026 (Show measurement, test conditions, and source)
- Source value
- 25.8
- Score
- 25.8
- Metric
- HMMT February 2026
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
LiveCodeBench v6 (Show measurement, test conditions, and source)
- Source value
- 37.4
- Score
- 37.4
- Metric
- LiveCodeBench v6
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
GPQA Diamond (Show measurement, test conditions, and source)
- Source value
- 27.3
- Score
- 27.3
- Metric
- GPQA Diamond
- Unit
- %
- Category
- source-specific
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- IFM model card
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Vendor-reported result. IFM-reported result for the exact K2 Horizon 0.9B card entry; no test-specific repeat count or harness is disclosed alongside these rows.
AutomationBench-AA (Show measurement, test conditions, and source)
- Source value
- 0.2038009788550524
- Score
- 0.2038009788550524
- Metric
- success share
- Unit
- %
- Category
- independent-evaluator
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis model profile
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Independent evaluation. Artificial Analysis's independent measurement of the variant explicitly labelled the reasoning version on its profile. The component belongs to Intelligence Index v4.3.2; the embedded model object provides the exact raw value.
- Context
- The public comparison table rounds normalized shares for display; the stored score preserves the raw model-object value multiplied by 100. This observation belongs to the exact Artificial Analysis reasoning-version profile and is not a cross-model cohort reconstructed here.
Terminal-Bench 4.0 (Show measurement, test conditions, and source)
- Source value
- 0
- Score
- 0
- Metric
- success share
- Unit
- %
- Benchmark version
- 4.0
- Category
- independent-evaluator
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis model profile
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Independent evaluation. Artificial Analysis's independent measurement of the variant explicitly labelled the reasoning version on its profile. The component belongs to Intelligence Index v4.3.2; the embedded model object provides the exact raw value.
- Context
- The public comparison table rounds normalized shares for display; the stored score preserves the raw model-object value multiplied by 100. This observation belongs to the exact Artificial Analysis reasoning-version profile and is not a cross-model cohort reconstructed here.
SciCode (Show measurement, test conditions, and source)
- Source value
- 6.94444444444444
- Score
- 6.94444444444444
- Metric
- success share
- Unit
- %
- Category
- independent-evaluator
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis model profile
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Independent evaluation. Artificial Analysis's independent measurement of the variant explicitly labelled the reasoning version on its profile. The component belongs to Intelligence Index v4.3.2; the embedded model object provides the exact raw value.
- Context
- The public comparison table rounds normalized shares for display; the stored score preserves the raw model-object value multiplied by 100. This observation belongs to the exact Artificial Analysis reasoning-version profile and is not a cross-model cohort reconstructed here.
Humanity's Last Exam (Show measurement, test conditions, and source)
- Source value
- 5.37534754402224
- Score
- 5.37534754402224
- Metric
- success share
- Unit
- %
- Category
- independent-evaluator
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis model profile
- Status
- active
- Retrieved at
- 2026-10-07
- Methodology
- Independent evaluation. Artificial Analysis's independent measurement of the variant explicitly labelled the reasoning version on its profile. The component belongs to Intelligence Index v4.3.2; the embedded model object provides the exact raw value.
- Context
- The public comparison table rounds normalized shares for display; the stored score preserves the raw model-object value multiplied by 100. This observation belongs to the exact Artificial Analysis reasoning-version profile and is not a cross-model cohort reconstructed here.
Page 1 of 2
Sources and data date
| Type | Evidence and data date |
|---|---|
| Input | Documented modalities. Retrieved October 7, 2026. |
| Output | Documented modalities. Retrieved October 7, 2026. |
| Research status | Research date October 7, 2026. 11 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not published | per-run-protocol-and-sample-counts Original wording The 0.9B results table does not disclose sample sizes, repeat counts, or full harness details. |
| Unresolved | excluded-non-target-checkpoints Original wording The pinned IFM family table contains values for intermediate checkpoint mid_2_final. Those weights are different from the final K2 Horizon 0.9B target profile. The midtraining values were excluded; final model-card results remain. |
| Additional source | IFM K2 Horizon release announcement (retrieved October 7, 2026) · IFM K2 Horizon announcement · Editorial description reviewed October 7, 2026 |
| Additional source | IFM model card (retrieved October 7, 2026) |
| Additional source | Artificial Analysis model profile (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |
| Additional source | Model metadata (retrieved October 7, 2026) |