fix model name returned by /api/generate
being different than the model name provided
This commit is contained in:
parent
045b855db9
commit
fa2f095bd9
2 changed files with 2 additions and 8 deletions
|
@ -545,7 +545,6 @@ type prediction struct {
|
||||||
const maxBufferSize = 512 * format.KiloByte
|
const maxBufferSize = 512 * format.KiloByte
|
||||||
|
|
||||||
type PredictOpts struct {
|
type PredictOpts struct {
|
||||||
Model string
|
|
||||||
Prompt string
|
Prompt string
|
||||||
Format string
|
Format string
|
||||||
CheckpointStart time.Time
|
CheckpointStart time.Time
|
||||||
|
@ -553,7 +552,6 @@ type PredictOpts struct {
|
||||||
}
|
}
|
||||||
|
|
||||||
type PredictResult struct {
|
type PredictResult struct {
|
||||||
Model string
|
|
||||||
CreatedAt time.Time
|
CreatedAt time.Time
|
||||||
TotalDuration time.Duration
|
TotalDuration time.Duration
|
||||||
LoadDuration time.Duration
|
LoadDuration time.Duration
|
||||||
|
@ -651,7 +649,6 @@ func (llm *llama) Predict(ctx context.Context, predict PredictOpts, fn func(Pred
|
||||||
|
|
||||||
if p.Content != "" {
|
if p.Content != "" {
|
||||||
fn(PredictResult{
|
fn(PredictResult{
|
||||||
Model: predict.Model,
|
|
||||||
CreatedAt: time.Now().UTC(),
|
CreatedAt: time.Now().UTC(),
|
||||||
Content: p.Content,
|
Content: p.Content,
|
||||||
})
|
})
|
||||||
|
@ -659,7 +656,6 @@ func (llm *llama) Predict(ctx context.Context, predict PredictOpts, fn func(Pred
|
||||||
|
|
||||||
if p.Stop {
|
if p.Stop {
|
||||||
fn(PredictResult{
|
fn(PredictResult{
|
||||||
Model: predict.Model,
|
|
||||||
CreatedAt: time.Now().UTC(),
|
CreatedAt: time.Now().UTC(),
|
||||||
TotalDuration: time.Since(predict.CheckpointStart),
|
TotalDuration: time.Since(predict.CheckpointStart),
|
||||||
|
|
||||||
|
|
|
@ -260,7 +260,7 @@ func GenerateHandler(c *gin.Context) {
|
||||||
}
|
}
|
||||||
|
|
||||||
resp := api.GenerateResponse{
|
resp := api.GenerateResponse{
|
||||||
Model: r.Model,
|
Model: req.Model,
|
||||||
CreatedAt: r.CreatedAt,
|
CreatedAt: r.CreatedAt,
|
||||||
Done: r.Done,
|
Done: r.Done,
|
||||||
Response: r.Content,
|
Response: r.Content,
|
||||||
|
@ -288,7 +288,6 @@ func GenerateHandler(c *gin.Context) {
|
||||||
|
|
||||||
// Start prediction
|
// Start prediction
|
||||||
predictReq := llm.PredictOpts{
|
predictReq := llm.PredictOpts{
|
||||||
Model: model.Name,
|
|
||||||
Prompt: prompt,
|
Prompt: prompt,
|
||||||
Format: req.Format,
|
Format: req.Format,
|
||||||
CheckpointStart: checkpointStart,
|
CheckpointStart: checkpointStart,
|
||||||
|
@ -985,7 +984,7 @@ func ChatHandler(c *gin.Context) {
|
||||||
loaded.expireTimer.Reset(sessionDuration)
|
loaded.expireTimer.Reset(sessionDuration)
|
||||||
|
|
||||||
resp := api.ChatResponse{
|
resp := api.ChatResponse{
|
||||||
Model: r.Model,
|
Model: req.Model,
|
||||||
CreatedAt: r.CreatedAt,
|
CreatedAt: r.CreatedAt,
|
||||||
Done: r.Done,
|
Done: r.Done,
|
||||||
Metrics: api.Metrics{
|
Metrics: api.Metrics{
|
||||||
|
@ -1007,7 +1006,6 @@ func ChatHandler(c *gin.Context) {
|
||||||
|
|
||||||
// Start prediction
|
// Start prediction
|
||||||
predictReq := llm.PredictOpts{
|
predictReq := llm.PredictOpts{
|
||||||
Model: model.Name,
|
|
||||||
Prompt: prompt,
|
Prompt: prompt,
|
||||||
Format: req.Format,
|
Format: req.Format,
|
||||||
CheckpointStart: checkpointStart,
|
CheckpointStart: checkpointStart,
|
||||||
|
|
Loading…
Reference in a new issue