Sulat.com
AI models
$10 off the fastest DeepSeek V4.1 Flash, Kimi K3 and GLM 5.3 from Synthetic
LLM Gateway logo

Model details

DeepSeek V4 Pro (Together AI)

DeepSeek V4 Pro 0813 is reported as an open-weight update to the V4 Pro line, positioned for demanding software-development workflows such as planning multi-file refactors and running commands. A third-party comparison associates the release with a substantial improvement on the Vals Index, reflecting progress in a model family intended to combine reasoning with practical agentic use.

The available evidence does not establish the architecture, training scale, parameter count, or detailed context behavior of the cataloged Together AI variant. Its practical fit should therefore be evaluated from the documented 0813 release only, with particular attention to coding, planning, and tool-oriented tasks rather than treating the variant as equivalent to the separately cataloged model.

LLM Gatewaytogether-ai/deepseek-v4-prodeepseek-thinking

Quick Info

Powered by
Provider
LLM Gateway
Model key
together-ai/deepseek-v4-pro
Release date
Apr 24, 2026
Last updated
Apr 24, 2026
Knowledge cutoff
2025-05
Input modalities
Output modalities
Capabilities

Cost

Input token cost
$1.32
Output token cost
$3.96

Limits

Output tokens
163,840 tokens
Context window
1,048,576 tokens

Transparent token rates

Compare DeepSeek V4 Pro (Together AI) pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about DeepSeek V4 Pro (Together AI)

LLM Gateway

CoverageBenchmark

According to a third-party technical guide, DeepSeek officially released the general-availability build of DeepSeek-V4-Pro (0813) on August 13, 2026, publishing open weights under the MIT license alongside cloud API endpoints. The guide frames the release as targeted at complex autonomous reasoning and repository-scale The same guide details the architecture as a Mixture-of-Experts model with 1.6 trillion total parameters and 49 billion active parameters per token, using fine-grained routing across 256 expert sub-networks with 8 active experts plus shared invariant experts. Multi-Head Latent Attention compresses the Key-Value cache f

Videos about DeepSeek V4 Pro (Together AI)

More models around DeepSeek V4 Pro (Together AI)