NVIDIA model product
NV-Embed-v2
NV-Embed-v2 is an embedding model from NVIDIA that is ranked on MTEB.
Updated Sep 8, 2026. Default version: NV-Embed-v2
LLMBoard Embedding Score79.2NV-Embed-v2
Coverage85%5 benchmark families
Context window32.8KTokens
Official input priceN/AOfficial price unavailable
NV-Embed-v2 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
NV-Embed-v2 Benchmark Results
Benchmark scores for NV-Embed-v2.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkMTEB English v2 Retrieval | Score62.84% | Rank13 | Participants212 | Percentile94.31% | EvidenceA | EvaluatedN/A |
| BenchmarkMTEB English v2 Reranking | Score49.61% | Rank23 | Participants241 | Percentile90.83% | EvidenceA | EvaluatedN/A |
| BenchmarkMTEB English v2 Classification | Score87.94% | Rank27 | Participants257 | Percentile89.84% | EvidenceA | EvaluatedN/A |
| BenchmarkMTEB English v2 Semantic Similarity | Score83.82% | Rank45 | Participants211 | Percentile79.05% | EvidenceA | EvaluatedN/A |
| BenchmarkMTEB English v2 Clustering | Score47.66% | Rank78 | Participants217 | Percentile64.35% | EvidenceA | EvaluatedN/A |
NV-Embed-v2 Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
NV-Embed-v2 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
NV-Embed-v2 Runtime Performance
Provider-specific output speed and catalog latency for NV-Embed-v2. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
NV-Embed-v2 Specifications
Technical details for the model's default version.
- Version
- NV-Embed-v2
- Released
- Sep 9, 2024
- Knowledge cutoff
- Unknown
- Parameters
- 7.9B
- Context window
- 32.8K
- Max output
- N/A
- Inputs
- text
- Outputs
- embedding
- Open weights
- Yes
- License
- cc-by-nc-4.0
NV-Embed-v2 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionNV-Embed-v2 | Released | LLMBoard79.21 | Parameters7.9B | Context32.8K | Max outputN/A | Open weightsYes | Licensecc-by-nc-4.0 |
Models similar to NV-Embed-v2
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
NV
NV-Embed-v1
NVIDIA
NV
nemotron-colembed-vl-8b-v2
NVIDIA
NV
nemotron-colembed-vl-4b-v2
NVIDIA
NV
llama-nemotron-colembed-vl-3b-v2
NVIDIA
NV
llama-nemoretriever-colembed-3b-v1
NVIDIA
NV
llama-nemoretriever-colembed-1b-v1
NVIDIA
What is NV-Embed-v2?
Key information about NV-Embed-v2 and its available data.
NV-Embed-v2 is an embedding model developed by NVIDIA. It is a ranked model on MTEB.
Data as of 2026-09-08.
FAQ
Common questions about NV-Embed-v2.
When was NV-Embed-v2 released?
NV-Embed-v2's default version was released on Sep 9, 2024.
How much does NV-Embed-v2 cost?
No official standard PAYG price is currently available for NV-Embed-v2.
Who created NV-Embed-v2?
NV-Embed-v2 was created by NVIDIA.
What is the context window for NV-Embed-v2?
The default version has a 32.8K token context window.
Is NV-Embed-v2 open weight?
Yes. The default version is marked as open weight under cc-by-nc-4.0.
How many API providers offer NV-Embed-v2?
No provider offering is currently linked to the default version.