✸
Website SDR chat
SalesA homepage chat that greets visitors, answers product questions, qualifies them, and books a demo — in real time, at high volume.
The routing recipe
Job typeGeneral chat / assistantPolicyFastest that passesConstraintsgroq
routes to
GPT-OSS 20B groq
groq:openai/gpt-oss-20b
87
match score
77.7
task fit
66.4
cost
96.4
speed
Why this model from benchmarks & capabilities
- "General chat / assistant" leans hardest on instruction; GPT-OSS 20B scores 78/100 there — #37 of 54.
- Strength on this class of work shows up on LMArena Elo, MT-Bench, IFEval, where OpenAI (open-weight) models are competitive.
- Cost: ~$0.244/Mtok blended ($0.07 in / $0.30 out). Speed: ~1000 tok/s on Groq LPU.
- Chosen over Llama 3.1 8B Instant for higher task-fit (77.7 vs 65.9) & faster.
| Capability the job needs | Weight | GPT-OSS 20B | Catalog rank |
|---|---|---|---|
| instruction | 0.80 | #37 of 54 | |
| knowledge | 0.60 | #40 of 54 | |
| speed | 0.60 | #2 of 54 |
Relevant benchmarksLMArena Elo MT-Bench IFEval
EconomicsCost: ~$0.244/Mtok blended ($0.07 in / $0.30 out).
Speed: ~1000 tok/s on Groq LPU. vs runner-upChosen over Llama 3.1 8B Instant for higher task-fit (77.7 vs 65.9) & faster. (Llama 3.1 8B Instant, score 85)
Speed: ~1000 tok/s on Groq LPU. vs runner-upChosen over Llama 3.1 8B Instant for higher task-fit (77.7 vs 65.9) & faster. (Llama 3.1 8B Instant, score 85)
Example result
Sample input ▾
Visitor: hey, do you help with SOC 2? we're a 40-person SaaS and an enterprise prospect is asking for our report.
Absolutely — SOC 2 is our bread and butter, and "an enterprise prospect is asking" is the #1 reason teams your size start. icompaas maps your AWS/Okta/GitHub stack to the SOC 2 controls, auto-collects evidence, and gets most 40-person SaaS teams audit-ready in weeks, not quarters. Quick one: are you going for Type I (point-in-time, faster) or Type II? I can grab you a 20-min walkthrough this week — what's your work email?
Test it on your own data