Ollama has added Gemma 4 renderer
The story so far
Ollama released the NVIDIA Nemotron 3.5 Bolt E4B visual language model based on Gemma 4, supporting a 128K token context window and DFlash speculative decoding for local dialogue, image understanding, and tool use. OpenAI updated, stating the model can optionally use the RENDERER gemma4-nemotron, supports DFlash sketches, exposes Gemma 4 hidden states and raw logits for speculative decoding, and provides independent reporting of safetensors target accuracy.
AI-written from the reports · Updated 26 days ago
Developments
2 developments
- 23 Oct, 22:12 · 1 reportsv0.40.10-rc0 Update gemma renderersOllama releases:v0.40.10
- 23 Oct, 22:12 · 1 reportsv0.40.10 Update gemma renderersOllama releases:v0.40.10
Coverage timeline
Follow the coverage to see every side of the story.
- Ollama releasesv0.40.10
NVIDIA Nemotron 3.5 Bolt E4B is a visual language model based on Gemma 4, supporting a 128K token context window, including DFlash speculative decoding, and available via ollama run nemotron-3.5-bolt:e4b.
- Ollama releasesv0.40.10
NVIDIA Nemotron 3.5 Bolt E4B is a visual language model based on Gemma 4, supporting a 128K token context window, including DFlash speculative decoding, used for local chat, image understanding, and tool use.