Kolibri 1
Kolibri 1 is available now for coding, rag, and agents with open-source and 1.05m context; evaluate it while provider pricing coverage matures.
- Family
- Kolibri
- Released
- 2026-10-03
- Context
- 1.05m
- Parameters
- 78B (3.46B activated)
- Architecture
- Mixture of Experts
- Knowledge cutoff
- 2026-06
- Specialization
- general
- Openness
- Open source
- License
- Apache 2.0OSI-approvedCommercial use: permitted
- Weights
- Available
- Code
- Unknown
- Training
- Multi-stage
No tracked provider token pricing is available yet.
About
Kolibri (Kolibri 1, HF Aleph-Alpha/Kolibri-1) is Aleph Alpha's open-weight English–German mixture-of-experts reasoning model, released 3 October 2026 and trained from scratch in Germany. It has 78B total parameters (78,103,074,560) with 3.46B active per token: 50 layers, 384 routed experts plus 1 shared, top-6 sigmoid routing, and 4:1 sliding-window-to-full GQA attention. Native context is 262,144 tokens. Because RoPE is applied only in the sliding-window layers, Aleph Alpha validates up to 1,048,576 tokens via a max_position_embeddings override, and recommends at most 262,144 for serving efficiency and complex tasks. It is text-in, text-out with controllable reasoning effort (low/medium/high/none) and Hermes-style tool calling.
Provider price ladder
No tracked provider token pricing is available for this model yet.
Capabilities
Benchmark peer barsfor Coding
Benchmark scores(11)
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| Google-Proof Q&A | 84.3 | diamondObserved 2026-10-03 | — | Source |
| MMLU PRO | 80.0 | MMLU-Pro CoT (EN)Observed 2026-10-03 | — | Source |
| AIME 2025 | 96.9 | AIME 2025 (EN)Observed 2026-10-03 | — | Source |
| AIME 2026 | 96.0 | AIME 2026 (EN)Observed 2026-10-03 | — | Source |
| LiveCodeBench | 85.9 | LiveCodeBench v6Observed 2026-10-03 | — | Source |
| HumanEval+ | 92.7 | HumanEval+Observed 2026-10-03 | — | Source |
| SWE-bench Verified | 66.4 | SWE-Bench Verified (Harbor)Observed 2026-10-03 | — | Source |
| Terminal-Bench 2.1 | 27.7 | TerminalBench 2.1 (Harbor)Observed 2026-10-03 | — | Source |
| BrowseComp | 29.4 | BrowseCompObserved 2026-10-03 | — | Source |
| BFCL | 61.4 | v4 (overall)Observed 2026-10-03 | — | Source |
| τ-bench | 94.7 | Tau2-Bench TelecomObserved 2026-10-03 | — | Source |
Migration checks
No linked migration route is available for this model yet.
API versions
Aleph-Alpha/Kolibri-1Aleph-Alpha/Kolibri-1-BF16Kolibri-1Kolibri 1KolibriNo tracked provider token pricing is available yet.