Leaderboard / Tilde
TildeOpen 30B
Tilde · TildeOpen · released 2026-02-04 · data checked 2026-11-01
TildeOpen 30B is a 30.7B-parameter dense model from Tilde (Europe), released 2026-02-04 under the CC-BY-4.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 21.7 GB of memory at 4-bit with an 8K-token context.
Base (pre-trained) model built for smaller European languages; not chat-tuned.
Small LM Index
—
Needs 2 core benchmarks (has 0)
Parameters
30.7B
Dense
Price per 1M tokens
—
No hosted API on OpenRouter; self-host
Context window
64K
tokens
Benchmarks
Each bar shows this model's score. The tick marks the best medium model on that benchmark.
No benchmark results for this release yet. Know of one? Send it.
Specifications
- Lab
- Tilde (LV · Europe)
- Architecture
- Dense transformer
- Total parameters
- 30.7B
- Active parameters
- All (dense)
- Size class
- Medium
- Context window
- 64K tokens
- Input modalities
- text
- Reasoning
- Instruct (no thinking)
- License
- CC-BY-4.0
- Openness
- Open weights
- Memory to run
- ~21.7 GB at 4-bit, 8K context (more sizes)
- Hugging Face
- 543 downloads (30 days) · 3 likes
Compare with
- TildeOpen 30B vs Gemma 4 31B (30.7B, index 70.1)
- TildeOpen 30B vs Kanana-2 30B-A3B Thinking (30.7B, index 50.7)
- TildeOpen 30B vs Qwen3 30B-A3B Thinking 2507 (30.5B, index 59.3)
Sources
Something wrong here? Report it.
Run it locally
No GGUF build from a well-known publisher yet.
vLLM (GPU server)
vllm serve TildeAI/TildeOpen-30b-64kFull-precision weights, for a server GPU.
Memory needed: weights, context cache and runtime buffers.
| Quantization | 4K context | 32K context | 64K context |
|---|---|---|---|
| 16-bit F16 | 64 GB | 71 GB | 78 GB |
| 8-bit Q8_0 | 35 GB | 42 GB | 49 GB |
| 4-bit Q4_K_M | 21 GB | 28 GB | 35 GB |