Kimi K3.1 Pro (batch)
- Input / output
- $0.24 / $8.40
- Providers
- 20
- Spread
- 3.2×
- 7 days
- −3.9%
- 30 days
- —
- Context
- 1M
Providers
| Provider | Input | Output | Blended | vs ref | Context | Uptime 1d | Price since |
|---|---|---|---|---|---|---|---|
RelaceCheapest fp4 | $0.24 | $8.40 | $2.28 | −17% | |||
Waferus mxfp4 | $0.253 | $8.80 | $2.39 | −13% | |||
| mxfp4·−40% discount | $1.20 | $6.00 | $2.40 | −13% | |||
Sail Researchfp4/us fp4 | $1.04 | $9.35 | $3.12 | +13% | |||
| fp4 | $1.40 | $9.35 | $3.39 | +23% | |||
| mxfp4 | $1.80 | $9.00 | $3.60 | +30% | |||
InferenceNetmxfp4/us mxfp4 | $1.80 | $9.00 | $3.60 | +30% | |||
| mxfp4 | $1.80 | $9.00 | $3.60 | +30% | |||
| bf16·−5% discount | $1.90 | $9.50 | $3.80 | +38% | |||
| fp4 | $1.75 | $10 | $3.81 | +38% | |||
| $2.00 | $9.49 | $3.87 | +40% | ||||
| $2.00 | $10 | $4.00 | +45% | ||||
| $2.00 | $10 | $4.00 | +45% | ||||
| $2.00 | $10 | $4.00 | +45% | ||||
| mxfp4 | $2.00 | $10 | $4.00 | +45% | |||
| mxfp4 | $2.00 | $10 | $4.00 | +45% | |||
| fp4 | $2.00 | $10 | $4.00 | +45% | |||
| fp4 | $2.00 | $10 | $4.00 | +45% | |||
| mxfp4 | $2.00 | $10 | $4.00 | +45% | |||
| $2.00 | $10 | $4.00 | +45% | ||||
| $2.00 | $10 | $4.00 | +45% | ||||
| fp8 | $0.975 | $14.62 | $4.387 | +59% | |||
InferenceNetfast mxfp4 | $2.70 | $13.50 | $5.40 | +96% | |||
InferenceNetus/fast mxfp4 | $2.70 | $13.50 | $5.40 | +96% | |||
Amazon Bedrockus-east-2 | $4.00 | $20 | $8.00 | +190% | |||
Moonshot AIhighspeed mxfp4 | $4.00 | $20 | $8.00 | +190% | |||
Amazon Bedrockus-east-1 | $4.40 | $22 | $8.80 | +219% |
- RelaceCheapest$2.28$0.24 in · $8.40 out−17% vs ref1M ctx100% uptimefp4
- us$2.39$0.253 in · $8.80 out−13% vs ref1M ctx100% uptimemxfp4
- $2.40$1.20 in · $6.00 out−13% vs ref1M ctx100% uptimemxfp4·−40% discount
- fp4/us$3.12$1.04 in · $9.35 out+13% vs ref1M ctx100% uptimefp4
- $3.39$1.40 in · $9.35 out+23% vs ref1M ctx100% uptimefp4
- $3.60$1.80 in · $9.00 out+30% vs ref1M ctx100% uptimemxfp4
- mxfp4/us$3.60$1.80 in · $9.00 out+30% vs ref1M ctx100% uptimemxfp4
- $3.60$1.80 in · $9.00 out+30% vs ref1M ctx100% uptimemxfp4
- $3.80$1.90 in · $9.50 out+38% vs ref1M ctx99.9% uptimebf16·−5% discount
- $3.81$1.75 in · $10 out+38% vs ref1M ctx100% uptimefp4
- $3.87$2.00 in · $9.49 out+40% vs ref1M ctx97.2% uptime
- $4.00$2.00 in · $10 out+45% vs ref1M ctx100% uptime
- $4.00$2.00 in · $10 out+45% vs ref1M ctx100% uptime
- $4.00$2.00 in · $10 out+45% vs ref1M ctx97.5% uptime
- $4.00$2.00 in · $10 out+45% vs ref1M ctx100% uptimemxfp4
- $4.00$2.00 in · $10 out+45% vs ref1M ctx99.9% uptimemxfp4
- $4.00$2.00 in · $10 out+45% vs ref1M ctx100% uptimefp4
- $4.00$2.00 in · $10 out+45% vs ref1M ctx99.9% uptimefp4
- $4.00$2.00 in · $10 out+45% vs ref1M ctx100% uptimemxfp4
- $4.00$2.00 in · $10 out+45% vs ref1M ctx99.9% uptime
- $4.00$2.00 in · $10 out+45% vs ref1M ctx99.8% uptime
- $4.387$0.975 in · $14.62 out+59% vs ref1M ctx98.7% uptimefp8
- fast$5.40$2.70 in · $13.50 out+96% vs ref1M ctx100% uptimemxfp4
- us/fast$5.40$2.70 in · $13.50 out+96% vs ref1M ctx100% uptimemxfp4
- us-east-2$8.00$4.00 in · $20 out+190% vs ref1M ctx100% uptime
- highspeed$8.00$4.00 in · $20 out+190% vs ref1M ctx98.4% uptimemxfp4
- us-east-1$8.80$4.40 in · $22 out+219% vs ref1M ctx100% uptime
Blended = (3 × input + output) / 4. Ref $2.76 = median of cheapest 20%.
Price history
USD per 1M tokens · each step is a price change
Prices in USD / 1M tokens.
Other pricing
| Offered by | Cache read | Cache write |
|---|---|---|
| Relace, Makora +11 | $0.20 | — |
| Wafer (us), Wafer +1 | $0.18 | — |
| Decart | $0.12 | — |
| Sail Research (fp4/us), Sail Research | $0.187 | — |
| InferenceNet, InferenceNet (mxfp4/us) | $0.189 | — |
| DeepInfra | $0.19 | — |
| Morph | $0.309 | — |
| InferenceNet (fast), InferenceNet (us/fast) | $0.27 | — |
| Amazon Bedrock (us-east-2) | $0.20 | $5.00 |
| Amazon Bedrock (us-east-1) | $0.22 | $5.50 |
Changes
- PriceCache read price,$0.353→$0.309−12.5%
- PriceOutput price,$16.67→$14.62−12.3%
- PriceInput price,$1.111→$0.975−12.2%
- PriceCache read price,$0.286→$0.353+23.4%
- PriceOutput price,$13.51→$16.67+23.3%
- PriceInput price,$0.901→$1.111+23.3%
- PriceCache read price,$0.353→$0.286−19.0%
- PriceOutput price,$16.66→$13.51−18.9%
- PriceInput price,$1.111→$0.901−18.9%
- PriceCache read price,$0.18→$0.189+5.0%
- PriceCache read price,$0.18→$0.189+5.0%
- PriceCache read price,$0.393→$0.353−10.2%
- PriceOutput price,$18.55→$16.66−10.2%
- PriceInput price,$1.236→$1.111−10.1%
- PriceCache read price,$0.202→$0.393+94.6%
About
The cheapest provider for Kimi K3.1 Pro (batch) is Relace at $2.28 per 1M tokens (blended), charging $0.24 per 1M input and $8.40 per 1M output tokens. 20 providers serve it; the most expensive charges 3.2× the typical price. Context window: 1M tokens. Prices tracked since Nov 12, 2026.
Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...
- Released
- Oct 23, 2026
- First tracked
- Nov 12, 2026
- Canonical slug
- moonshotai/kimi-k3.1-pro-20261023
- Weights
- moonshotai/Kimi-K3.1-Pro