Nemotron 3.5 Bolt 74B-A9B
NVIDIAOpen weightsReasoningProvisionalReleased Oct 15, 2026
Overall score
45.3
Rank 124 of 222 · 30.9–59.6
Context window
256K
256,000 tokens
Price per 1M tokens
— in · — out
No price listed
Blended price
—
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Reasoning72.9
- 2Agentic & tool use48.7
- 3Knowledge43.1
Benchmark results (7)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| Terminal-Bench 2.1 | 51.0 | #59 of 68 |
| BrowseComp | 66.8 | #35 of 50 |
| MCP Atlas | 52.2 | #35 of 42 |
| τ³-bench results | 17.0 | #20 of 23 |
| PinchBench | 87.8 | #3 of 7 |
| WebArena | 50.9 | #1 of 1 |
Knowledge
| HLE w/o tools | 25.3 | #37 of 43 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price | — | — | — |
NVIDIA publishes downloadable OpenMDW 1.1 weights for local inference. No first-party per-token hosted API tariff is established for this checkpoint; hardware costs are separate.
Speed
Measured by us through OpenRouter (median of trials).
Not available on OpenRouter, so we cannot measure it yet.
Related models
Same family
Similar score
- o3-pro45.4 · $35.00
- Ling 3.0 Flash45.1 · $0.032
- o345.4 · $3.50
- Gemma 4 26B A4B45.6 · $0.12
- Qwen3.5-35B-A3B45.8 · $0.35