Baseten: Models Intelligence, Performance & Price
This analysis is intended to support you in choosing the best model provided by Baseten for your use-case.
Most Intelligent
Kimi K3.1 Pro
DeepSeek V4.1 Flash 1017 (max)DeepSeek V4.1 Flash 1017 (max) (FAST)
GLM-5.3 (max)GLM-5.3 (max) (FAST)Intelligence Index
Total 22 models
Lowest Time per Task
Beam 1.0 (NVFP4)DeepSeek V4.1 Flash 1017 (max) (FAST)
Nemotron 3.5 Super VL
Muse Spark Open (xhigh)Kimi K3 (low)Time per Intelligence Index task (minutes)
Total 22 models
Lowest Price
DeepSeek V4 Flash 0731 (max)GLM-5.3-FlashNemotron 3.5 Super VL
gpt-oss-120b (high)gpt-oss-120b (low)Blended price (per 1M tokens)
Total 22 models
Baseten offers 22 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For intelligence, the top models on Baseten are Kimi K3.1 Pro (46), DeepSeek V4.1 Flash 1017 (max) (33), and DeepSeek V4.1 Flash 1017 (max) (FAST) (33).
- For time per task, Beam 1.0 (NVFP4) (1.4 min), DeepSeek V4.1 Flash 1017 (max) (FAST) (2.8 min), and Nemotron 3.5 Super VL (2.9 min) finish an Intelligence Index task fastest on Baseten. Time per task varies significantly across models, with the slowest taking 3.4x longer than the quickest.
- For latency, DeepSeek V4.1 Flash (non-reasoning) (FAST) (0.31s), gpt-oss-120b (low) (0.70s), and GLM-5.2 (non-reasoning) (3.71s) offer the lowest time to first answer token.
- For pricing, DeepSeek V4 Flash 0731 (max) ($0.07), GLM-5.3-Flash ($0.10), and Nemotron 3.5 Super VL ($0.12) offer the lowest blended prices per 1M tokens.
- For context window size, Kimi K3.1 Pro (1M), DeepSeek V4.1 Flash 1017 (max) (FAST) (1M), and GLM-5.3 (max) (1M) support the largest context windows on Baseten.
Intelligence
Artificial Analysis Intelligence Index
Evaluations
Agentic knowledge work, Elo
Knowledge
1 - hallucination rate
Agentic SaaS workflows
Agentic scientific research workflows in a terminal
Agentic software engineering in open-source repositories
Reasoning & knowledge
Professional document reasoning, All-pass
Agentic coding & terminal use
Physics reasoning
Quantitative analysis on spreadsheets & documents
Legal agentic work, Hallucination-Gated All-Pass Rate
Agentic business operations
Agentic real-world work tasks, Elo
Agentic interactive 3D experiences, Elo
Product and brand videos designed in code, Elo
Kubernetes incident root-cause analysis
Medical long context reasoning
Visual reasoning
Rich, varied, well-crafted language
Well-structured, scannable, answer-first
Truthful, acknowledges uncertainty, non-sycophantic
Use of filler and generic language
Intelligence Index vs. Price
Pricing
Cost per Intelligence Index Task
Intelligence Index vs. Price
Speed
Average of the single-request and multi-turn agentic workloads measured
Time per Intelligence Index Task
Quality
Endpoint Accuracy Index
Latency
Average of the single-request and multi-turn agentic workloads measured
Time to First Answer Token
Reliability
Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured
Endpoint Error Rate
Caching
Cache Hit Rate
Cost per Task vs. Cache Hit Rate
Further Analysis | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Kimi K3.1 Pro | 46 | $3.26 | 7.2 | 33.01 | 0.0% | 1M | Open | |||
DeepSeek V4.1 Flash 1017 (max) | 33 | $1.98 | 15.8 | 46.90 | 5.5% | 1M | Open | |||
DeepSeek V4.1 Flash 1017 (max) (FAST) | 33 | $2.32 | 2.8 | 9.21 | 0.7% | 1M | Open | |||
GLM-5.3 (max) | 32 | $3.45 | 19.0 | 47.38 | 4.9% | 1M | Open | |||
GLM-5.3 (max) (FAST) | 32 | $9.52 | 8.9 | 20.30 | 6.1% | 1M | Open | |||
GLM-5.3-Flash | 28 | $1.74 | 19.6 | 65.41 | 0.0% | 1M | Open | |||
Muse Spark Open (xhigh) | 25 | $2.81 | 4.8 | 7.89 | 4.0% | 1M | Open | |||
DeepSeek V4 Pro 0813 (max) | 23 | $4.95 | 15.6 | 45.95 | 5.6% | 1M | Open | |||
Beam 1.0 (NVFP4) | 18 | $0.45 | 1.3 | 3.75 | 4.3% | 1M | Open | |||
Kimi K3 (low) | 17 | $3.44 | 5.0 | 3.42 | 0.0% | 1M | Open | |||
Nemotron 3.5 Super VL | 14 | $0.31 | 2.9 | 6.83 | 4.9% | 1M | Open | |||
gpt-oss-120b (high) | 4 | $0.24 | 6.0 | 14.64 | 5.2% | 128k | Open | |||
Kimi K2.6 | -- | -- | -- | -- | -- | 262k | Open | |||
Kimi K2.6 (non-reasoning) | -- | -- | -- | -- | -- | 262k | Open | |||
DeepSeek V4 Pro (high) | -- | -- | -- | -- | -- | 1M | Open | |||
DeepSeek V4 Pro (non-reasoning) | -- | -- | -- | -- | -- | 1M | Open | |||
DeepSeek V4.1 Flash (non-reasoning) (FAST) | -- | -- | -- | 0.31 | 0.0% | 1M | Open | |||
DeepSeek V4 Pro (max) | -- | -- | -- | -- | -- | 1M | Open | |||
gpt-oss-120b (low) | -- | -- | -- | 0.64 | 4.0% | 128k | Open | |||
DeepSeek V4.1 Flash (max) | -- | -- | -- | 38.02 | 0.0% | 1M | Open | |||
DeepSeek V4.1 Flash (max) (FAST) | -- | -- | -- | 10.81 | 0.0% | 1M | Open | |||
GLM-5.2 (non-reasoning) (FAST) | -- | -- | -- | 8.10 | 0.0% | 524k | Open | |||
GLM-5.2 (non-reasoning) | -- | -- | -- | 3.14 | 0.0% | 1M | Open | |||
GLM-5.2 (max) (FAST) | -- | -- | -- | 12.77 | 0.0% | 524k | Open | |||
GLM-5.2 (max) | -- | -- | -- | 31.99 | 2.5% | 1M | Open | |||
Kimi K3 (max) | -- | -- | -- | 23.53 | 1.9% | 1M | Open | |||
DeepSeek V4 Flash 0731 (max) | -- | -- | -- | 11.14 | 2.1% | 1M | Open | |||
Key definitions
Frequently Asked Questions
Common questions about Baseten
Baseten offers 22 models that we track: Kimi K3.1 Pro, DeepSeek V4.1 Flash 1017 (max), DeepSeek V4.1 Flash 1017 (max) (FAST), GLM-5.3 (max), GLM-5.3 (max) (FAST), GLM-5.3-Flash, Muse Spark Open (xhigh), DeepSeek V4 Pro 0813 (max), Beam 1.0 (NVFP4), Kimi K3 (low), Nemotron 3.5 Super VL, gpt-oss-120b (high), DeepSeek V4.1 Flash (non-reasoning) (FAST), gpt-oss-120b (low), DeepSeek V4.1 Flash (max), DeepSeek V4.1 Flash (max) (FAST), GLM-5.2 (non-reasoning) (FAST), GLM-5.2 (non-reasoning), GLM-5.2 (max) (FAST), GLM-5.2 (max), Kimi K3 (max), and DeepSeek V4 Flash 0731 (max).
The most intelligent model available on Baseten is Kimi K3.1 Pro with an Intelligence Index score of 46.
The model with the lowest time per task on Baseten is Beam 1.0 (NVFP4) at 1.4 minutes per Intelligence Index task. Time per task is the output tokens per task divided by end-to-end output speed, so lower means a task finishes sooner.
The model with the lowest time to first answer token on Baseten is DeepSeek V4.1 Flash (non-reasoning) (FAST) at 0.31s. Lower latency means faster initial response time.
The most affordable model on Baseten by blended price is DeepSeek V4 Flash 0731 (max) at $0.0716 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on Baseten vary up to 32.3x across models, from $0.0716 per 1M tokens for DeepSeek V4 Flash 0731 (max) to $2.31 per 1M tokens for Kimi K3 (max).
Yes, Baseten offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 22 models on Baseten support JSON mode for structured output.
Yes, all 22 models on Baseten support function calling (tool use).
Yes, Baseten offers 19 reasoning models: Kimi K3.1 Pro, DeepSeek V4.1 Flash 1017 (max), DeepSeek V4.1 Flash 1017 (max) (FAST), GLM-5.3 (max), GLM-5.3 (max) (FAST), GLM-5.3-Flash, Muse Spark Open (xhigh), DeepSeek V4 Pro 0813 (max), Beam 1.0 (NVFP4), Kimi K3 (low), Nemotron 3.5 Super VL, gpt-oss-120b (high), gpt-oss-120b (low), DeepSeek V4.1 Flash (max), DeepSeek V4.1 Flash (max) (FAST), GLM-5.2 (max) (FAST), GLM-5.2 (max), Kimi K3 (max), and DeepSeek V4 Flash 0731 (max). Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, all 22 models on Baseten are open weight models.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
When choosing a model on Baseten, consider: intelligence (for quality-sensitive tasks), time per task (how quickly the model completes an Intelligence Index task), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.