Inference platformswww.together.ai13 published changes
Users relying on Llama Guard 4 12B ($0.20/token) through Together AI will need to find alternative providers or models for content moderation/safety use cases.
Developers can now evaluate the cost-performance tradeoff between GLM-5.3 and its faster Flash variant, with Flash offering 17x lower cost at the expense of ~5.6 points in pass@1 performance.
www.together.ai · also seen at 1 more source
New model option available to users; GLM-5.2 remains available as an alternative.
Marketing messaging changed from Series C funding announcement to model comparison focus, suggesting shift in promotional strategy
Users can now access newer DeepSeek-V4 model variants (Flash 0731, Flash) at $6.00-$15.00 pricing, replacing older R1 and V3 series offerings
www.together.ai · also seen at 1 more source
Users now have access to a new DeepSeek V4 Pro 0813 variant with lower input pricing ($1.32 vs $1.74) and higher output pricing ($3.96 vs $3.48), with reduced cached token pricing ($0.13 vs $0.20).
New model option available to users with standard and cached pricing tiers
Developers can now compare cost-effectiveness of different LLMs for coding tasks; highlights DeepSeek's superior price performance for coding workloads
New model option available for users, potential cost consideration for workloads using this model
Users can now access DeepSeek V4 Flash 0731 model through Together AI's platform with input cost of $0.14/MTok, cached input cost of $0.03/MTok, and output cost of $0.28/MTok
New model option available to users of Together AI's API for inference tasks.
www.together.ai · also seen at 1 more source
Developers can now use Kimi K3 model on Together AI platform with $3.00 input cost and $15.00 output cost, plus a $0.30 cached option for inputs.
Users can now access comparative analysis of Kimi K3 and Claude Fable 5 models on DeepSWE benchmark, helping inform model selection decisions based on pass rates and cost-efficiency metrics.