The Known Good Updated 8 Oct 2026 Subscribe
Providers· First-party endpoint· 16 tracked models

Reka — pricing & performance

Compare models All providers Data sources
Models served
16
of 1,191 tracked models
Mean output speed
66
tokens per second, averaged over
16 measured endpoints
Mean latency
1.0s
time to first token, averaged over
16 measured endpoints
Cheapest blended
$0.095
per 1M tokens at a 3:1
input:output blend
Output speed on Reka

Output tokens per second for all 16 models on Reka with published throughput telemetry, the most recent figure OpenRouter publishes for each host. Higher is better. We do not run these measurements ourselves.

⇩
149
128
107
98
90
72
66
62
61
55
46
44
The Known Good
16 models on Reka
Full leaderboard →
Model Creator Input $/M Output $/M Blended Speed TTFT Context
DeepSeek: DeepSeek V4 Flash 0423 DeepSeek $0.13 $0.26 $0.16 20 4.2s 1.05M
DeepSeek: DeepSeek V4 Flash 0731 DeepSeek $0.021 $0.53 $0.15 98 0.6s 262k
DeepSeek: DeepSeek V4 Pro 0423 DeepSeek $1.05 $10.50 $3.41 44 0.6s 1.05M
DeepSeek: DeepSeek V4 Pro 0813 DeepSeek $1.12 $3.37 $1.68 62 1.0s 1.05M
DeepSeek: DeepSeek V4.1 Flash DeepSeek $0.10 $0.84 $0.29 46 0.6s 1.05M
Google: Gemma 4 26B A4B Google $0.060 $0.20 $0.095 12 2.0s 262k
Google: Gemma 4 31B Google $0.080 $0.30 $0.14 15 0.6s 262k
Qwen: Qwen3.6 27B Alibaba $0.29 $1.95 $0.70 72 0.5s 262k
Qwen: Qwen3.6 35B A3B Alibaba $0.10 $1.00 $0.33 107 0.4s 262k
Qwen: Qwen3.8 27B Alibaba $0.024 $2.17 $0.56 61 0.3s 262k
Reka Edge Reka AI $0.10 $0.10 $0.10 128 0.5s 16k
Reka Flash 3 Reka AI $0.10 $0.20 $0.12 90 1.0s 66k
Tencent: Hy-MT2-30B-A3B Tencent $0.070 $0.28 $0.12 66 0.2s 33k
Z.ai: GLM 5.2 Z.ai $0.80 $4.00 $1.60 149 1.6s 1.05M
Z.ai: GLM 5.3 Z.ai $0.14 $4.20 $1.16 55 0.5s 1.05M
Z.ai: GLM 5.3 Flash Z.ai $0.060 $1.00 $0.29 24 1.2s 262k
The Known Good
Source. Prices, endpoint availability and throughput for Reka are ingested from OpenRouter; model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint — it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this site; see methodology. Quality scores are not provider-specific: a model scores the same wherever it is hosted, so quality lives on the model pages.